Merge origin/main into fix/structured-retry-classification-main

Made-with: Cursor

tests/agent/test_consolidate_offset.py

@@ -506,7 +506,7 @@ class TestNewCommandArchival:
 
     @pytest.mark.asyncio
     async def test_new_clears_session_immediately_even_if_archive_fails(self, tmp_path: Path) -> None:
-        """/new clears session immediately; archive_messages retries until raw dump."""
+        """/new clears session immediately; archive is fire-and-forget."""
         from nanobot.bus.events import InboundMessage
 
         loop = self._make_loop(tmp_path)
@@ -518,12 +518,12 @@ class TestNewCommandArchival:
 
         call_count = 0
 
-        async def _failing_consolidate(_messages) -> bool:
+        async def _failing_summarize(_messages) -> bool:
             nonlocal call_count
             call_count += 1
             return False
 
-        loop.memory_consolidator.consolidate_messages = _failing_consolidate  # type: ignore[method-assign]
+        loop.consolidator.archive = _failing_summarize  # type: ignore[method-assign]
 
         new_msg = InboundMessage(channel="cli", sender_id="user", chat_id="test", content="/new")
         response = await loop._process_message(new_msg)
@@ -535,7 +535,7 @@ class TestNewCommandArchival:
         assert len(session_after.messages) == 0
 
         await loop.close_mcp()
-        assert call_count == 3  # retried up to raw-archive threshold
+        assert call_count == 1
 
     @pytest.mark.asyncio
     async def test_new_archives_only_unconsolidated_messages(self, tmp_path: Path) -> None:
@@ -551,12 +551,12 @@ class TestNewCommandArchival:
 
         archived_count = -1
 
-        async def _fake_consolidate(messages) -> bool:
+        async def _fake_summarize(messages) -> bool:
             nonlocal archived_count
             archived_count = len(messages)
             return True
 
-        loop.memory_consolidator.consolidate_messages = _fake_consolidate  # type: ignore[method-assign]
+        loop.consolidator.archive = _fake_summarize  # type: ignore[method-assign]
 
         new_msg = InboundMessage(channel="cli", sender_id="user", chat_id="test", content="/new")
         response = await loop._process_message(new_msg)
@@ -578,10 +578,10 @@ class TestNewCommandArchival:
             session.add_message("assistant", f"resp{i}")
         loop.sessions.save(session)
 
-        async def _ok_consolidate(_messages) -> bool:
+        async def _ok_summarize(_messages) -> bool:
             return True
 
-        loop.memory_consolidator.consolidate_messages = _ok_consolidate  # type: ignore[method-assign]
+        loop.consolidator.archive = _ok_summarize  # type: ignore[method-assign]
 
         new_msg = InboundMessage(channel="cli", sender_id="user", chat_id="test", content="/new")
         response = await loop._process_message(new_msg)
@@ -604,12 +604,12 @@ class TestNewCommandArchival:
 
         archived = asyncio.Event()
 
-        async def _slow_consolidate(_messages) -> bool:
+        async def _slow_summarize(_messages) -> bool:
             await asyncio.sleep(0.1)
             archived.set()
             return True
 
-        loop.memory_consolidator.consolidate_messages = _slow_consolidate  # type: ignore[method-assign]
+        loop.consolidator.archive = _slow_summarize  # type: ignore[method-assign]
 
         new_msg = InboundMessage(channel="cli", sender_id="user", chat_id="test", content="/new")
         await loop._process_message(new_msg)

tests/agent/test_consolidator.py

@@ -0,0 +1,78 @@
+"""Tests for the lightweight Consolidator — append-only to HISTORY.md."""
+
+import pytest
+import asyncio
+from unittest.mock import AsyncMock, MagicMock, patch
+
+from nanobot.agent.memory import Consolidator, MemoryStore
+
+
+@pytest.fixture
+def store(tmp_path):
+    return MemoryStore(tmp_path)
+
+
+@pytest.fixture
+def mock_provider():
+    p = MagicMock()
+    p.chat_with_retry = AsyncMock()
+    return p
+
+
+@pytest.fixture
+def consolidator(store, mock_provider):
+    sessions = MagicMock()
+    sessions.save = MagicMock()
+    return Consolidator(
+        store=store,
+        provider=mock_provider,
+        model="test-model",
+        sessions=sessions,
+        context_window_tokens=1000,
+        build_messages=MagicMock(return_value=[]),
+        get_tool_definitions=MagicMock(return_value=[]),
+        max_completion_tokens=100,
+    )
+
+
+class TestConsolidatorSummarize:
+    async def test_summarize_appends_to_history(self, consolidator, mock_provider, store):
+        """Consolidator should call LLM to summarize, then append to HISTORY.md."""
+        mock_provider.chat_with_retry.return_value = MagicMock(
+            content="User fixed a bug in the auth module."
+        )
+        messages = [
+            {"role": "user", "content": "fix the auth bug"},
+            {"role": "assistant", "content": "Done, fixed the race condition."},
+        ]
+        result = await consolidator.archive(messages)
+        assert result is True
+        entries = store.read_unprocessed_history(since_cursor=0)
+        assert len(entries) == 1
+
+    async def test_summarize_raw_dumps_on_llm_failure(self, consolidator, mock_provider, store):
+        """On LLM failure, raw-dump messages to HISTORY.md."""
+        mock_provider.chat_with_retry.side_effect = Exception("API error")
+        messages = [{"role": "user", "content": "hello"}]
+        result = await consolidator.archive(messages)
+        assert result is True  # always succeeds
+        entries = store.read_unprocessed_history(since_cursor=0)
+        assert len(entries) == 1
+        assert "[RAW]" in entries[0]["content"]
+
+    async def test_summarize_skips_empty_messages(self, consolidator):
+        result = await consolidator.archive([])
+        assert result is False
+
+
+class TestConsolidatorTokenBudget:
+    async def test_prompt_below_threshold_does_not_consolidate(self, consolidator):
+        """No consolidation when tokens are within budget."""
+        session = MagicMock()
+        session.last_consolidated = 0
+        session.messages = [{"role": "user", "content": "hi"}]
+        session.key = "test:key"
+        consolidator.estimate_session_prompt_tokens = MagicMock(return_value=(100, "tiktoken"))
+        consolidator.archive = AsyncMock(return_value=True)
+        await consolidator.maybe_consolidate_by_tokens(session)
+        consolidator.archive.assert_not_called()

tests/agent/test_context_prompt_cache.py

@@ -47,6 +47,19 @@ def test_system_prompt_stays_stable_when_clock_changes(tmp_path, monkeypatch) ->
     assert prompt1 == prompt2
 
 
+def test_system_prompt_reflects_current_dream_memory_contract(tmp_path) -> None:
+    workspace = _make_workspace(tmp_path)
+    builder = ContextBuilder(workspace)
+
+    prompt = builder.build_system_prompt()
+
+    assert "memory/history.jsonl" in prompt
+    assert "automatically managed by Dream" in prompt
+    assert "do not edit directly" in prompt
+    assert "memory/HISTORY.md" not in prompt
+    assert "write important facts here" not in prompt
+
+
 def test_runtime_context_is_separate_untrusted_user_message(tmp_path) -> None:
     """Runtime metadata should be merged with the user message."""
     workspace = _make_workspace(tmp_path)

tests/agent/test_dream.py

@@ -0,0 +1,97 @@
+"""Tests for the Dream class — two-phase memory consolidation via AgentRunner."""
+
+import pytest
+
+from unittest.mock import AsyncMock, MagicMock
+
+from nanobot.agent.memory import Dream, MemoryStore
+from nanobot.agent.runner import AgentRunResult
+
+
+@pytest.fixture
+def store(tmp_path):
+    s = MemoryStore(tmp_path)
+    s.write_soul("# Soul\n- Helpful")
+    s.write_user("# User\n- Developer")
+    s.write_memory("# Memory\n- Project X active")
+    return s
+
+
+@pytest.fixture
+def mock_provider():
+    p = MagicMock()
+    p.chat_with_retry = AsyncMock()
+    return p
+
+
+@pytest.fixture
+def mock_runner():
+    return MagicMock()
+
+
+@pytest.fixture
+def dream(store, mock_provider, mock_runner):
+    d = Dream(store=store, provider=mock_provider, model="test-model", max_batch_size=5)
+    d._runner = mock_runner
+    return d
+
+
+def _make_run_result(
+    stop_reason="completed",
+    final_content=None,
+    tool_events=None,
+    usage=None,
+):
+    return AgentRunResult(
+        final_content=final_content or stop_reason,
+        stop_reason=stop_reason,
+        messages=[],
+        tools_used=[],
+        usage={},
+        tool_events=tool_events or [],
+    )
+
+
+class TestDreamRun:
+    async def test_noop_when_no_unprocessed_history(self, dream, mock_provider, mock_runner, store):
+        """Dream should not call LLM when there's nothing to process."""
+        result = await dream.run()
+        assert result is False
+        mock_provider.chat_with_retry.assert_not_called()
+        mock_runner.run.assert_not_called()
+
+    async def test_calls_runner_for_unprocessed_entries(self, dream, mock_provider, mock_runner, store):
+        """Dream should call AgentRunner when there are unprocessed history entries."""
+        store.append_history("User prefers dark mode")
+        mock_provider.chat_with_retry.return_value = MagicMock(content="New fact")
+        mock_runner.run = AsyncMock(return_value=_make_run_result(
+            tool_events=[{"name": "edit_file", "status": "ok", "detail": "memory/MEMORY.md"}],
+        ))
+        result = await dream.run()
+        assert result is True
+        mock_runner.run.assert_called_once()
+        spec = mock_runner.run.call_args[0][0]
+        assert spec.max_iterations == 10
+        assert spec.fail_on_tool_error is False
+
+    async def test_advances_dream_cursor(self, dream, mock_provider, mock_runner, store):
+        """Dream should advance the cursor after processing."""
+        store.append_history("event 1")
+        store.append_history("event 2")
+        mock_provider.chat_with_retry.return_value = MagicMock(content="Nothing new")
+        mock_runner.run = AsyncMock(return_value=_make_run_result())
+        await dream.run()
+        assert store.get_last_dream_cursor() == 2
+
+    async def test_compacts_processed_history(self, dream, mock_provider, mock_runner, store):
+        """Dream should compact history after processing."""
+        store.append_history("event 1")
+        store.append_history("event 2")
+        store.append_history("event 3")
+        mock_provider.chat_with_retry.return_value = MagicMock(content="Nothing new")
+        mock_runner.run = AsyncMock(return_value=_make_run_result())
+        await dream.run()
+        # After Dream, cursor is advanced and 3, compact keeps last max_history_entries
+        entries = store.read_unprocessed_history(since_cursor=0)
+        assert all(e["cursor"] > 0 for e in entries)
+

tests/agent/test_git_store.py

@@ -0,0 +1,234 @@
+"""Tests for GitStore — git-backed version control for memory files."""
+
+import pytest
+from pathlib import Path
+
+from nanobot.utils.gitstore import GitStore, CommitInfo
+
+
+TRACKED = ["SOUL.md", "USER.md", "memory/MEMORY.md"]
+
+
+@pytest.fixture
+def git(tmp_path):
+    """Uninitialized GitStore."""
+    return GitStore(tmp_path, tracked_files=TRACKED)
+
+
+@pytest.fixture
+def git_ready(git):
+    """Initialized GitStore with one initial commit."""
+    git.init()
+    return git
+
+
+class TestInit:
+    def test_not_initialized_by_default(self, git, tmp_path):
+        assert not git.is_initialized()
+        assert not (tmp_path / ".git").is_dir()
+
+    def test_init_creates_git_dir(self, git, tmp_path):
+        assert git.init()
+        assert (tmp_path / ".git").is_dir()
+
+    def test_init_idempotent(self, git_ready):
+        assert not git_ready.init()
+
+    def test_init_creates_gitignore(self, git_ready):
+        gi = git_ready._workspace / ".gitignore"
+        assert gi.exists()
+        content = gi.read_text(encoding="utf-8")
+        for f in TRACKED:
+            assert f"!{f}" in content
+
+    def test_init_touches_tracked_files(self, git_ready):
+        for f in TRACKED:
+            assert (git_ready._workspace / f).exists()
+
+    def test_init_makes_initial_commit(self, git_ready):
+        commits = git_ready.log()
+        assert len(commits) == 1
+        assert "init" in commits[0].message
+
+
+class TestBuildGitignore:
+    def test_subdirectory_dirs(self, git):
+        content = git._build_gitignore()
+        assert "!memory/\n" in content
+        for f in TRACKED:
+            assert f"!{f}\n" in content
+        assert content.startswith("/*\n")
+
+    def test_root_level_files_no_dir_entries(self, tmp_path):
+        gs = GitStore(tmp_path, tracked_files=["a.md", "b.md"])
+        content = gs._build_gitignore()
+        assert "!a.md\n" in content
+        assert "!b.md\n" in content
+        dir_lines = [l for l in content.split("\n") if l.startswith("!") and l.endswith("/")]
+        assert dir_lines == []
+
+
+class TestAutoCommit:
+    def test_returns_none_when_not_initialized(self, git):
+        assert git.auto_commit("test") is None
+
+    def test_commits_file_change(self, git_ready):
+        (git_ready._workspace / "SOUL.md").write_text("updated", encoding="utf-8")
+        sha = git_ready.auto_commit("update soul")
+        assert sha is not None
+        assert len(sha) == 8
+
+    def test_returns_none_when_no_changes(self, git_ready):
+        assert git_ready.auto_commit("no change") is None
+
+    def test_commit_appears_in_log(self, git_ready):
+        ws = git_ready._workspace
+        (ws / "SOUL.md").write_text("v2", encoding="utf-8")
+        sha = git_ready.auto_commit("update soul")
+        commits = git_ready.log()
+        assert len(commits) == 2
+        assert commits[0].sha == sha
+
+    def test_does_not_create_empty_commits(self, git_ready):
+        git_ready.auto_commit("nothing 1")
+        git_ready.auto_commit("nothing 2")
+        assert len(git_ready.log()) == 1  # only init commit
+
+
+class TestLog:
+    def test_empty_when_not_initialized(self, git):
+        assert git.log() == []
+
+    def test_newest_first(self, git_ready):
+        ws = git_ready._workspace
+        for i in range(3):
+            (ws / "SOUL.md").write_text(f"v{i}", encoding="utf-8")
+            git_ready.auto_commit(f"commit {i}")
+
+        commits = git_ready.log()
+        assert len(commits) == 4  # init + 3
+        assert "commit 2" in commits[0].message
+        assert "init" in commits[-1].message
+
+    def test_max_entries(self, git_ready):
+        ws = git_ready._workspace
+        for i in range(10):
+            (ws / "SOUL.md").write_text(f"v{i}", encoding="utf-8")
+            git_ready.auto_commit(f"c{i}")
+        assert len(git_ready.log(max_entries=3)) == 3
+
+    def test_commit_info_fields(self, git_ready):
+        c = git_ready.log()[0]
+        assert isinstance(c, CommitInfo)
+        assert len(c.sha) == 8
+        assert c.timestamp
+        assert c.message
+
+
+class TestDiffCommits:
+    def test_empty_when_not_initialized(self, git):
+        assert git.diff_commits("a", "b") == ""
+
+    def test_diff_between_two_commits(self, git_ready):
+        ws = git_ready._workspace
+        (ws / "SOUL.md").write_text("original", encoding="utf-8")
+        git_ready.auto_commit("v1")
+        (ws / "SOUL.md").write_text("modified", encoding="utf-8")
+        git_ready.auto_commit("v2")
+
+        commits = git_ready.log()
+        diff = git_ready.diff_commits(commits[1].sha, commits[0].sha)
+        assert "modified" in diff
+
+    def test_invalid_sha_returns_empty(self, git_ready):
+        assert git_ready.diff_commits("deadbeef", "cafebabe") == ""
+
+
+class TestFindCommit:
+    def test_finds_by_prefix(self, git_ready):
+        ws = git_ready._workspace
+        (ws / "SOUL.md").write_text("v2", encoding="utf-8")
+        sha = git_ready.auto_commit("v2")
+        found = git_ready.find_commit(sha[:4])
+        assert found is not None
+        assert found.sha == sha
+
+    def test_returns_none_for_unknown(self, git_ready):
+        assert git_ready.find_commit("deadbeef") is None
+
+
+class TestShowCommitDiff:
+    def test_returns_commit_with_diff(self, git_ready):
+        ws = git_ready._workspace
+        (ws / "SOUL.md").write_text("content", encoding="utf-8")
+        sha = git_ready.auto_commit("add content")
+        result = git_ready.show_commit_diff(sha)
+        assert result is not None
+        commit, diff = result
+        assert commit.sha == sha
+        assert "content" in diff
+
+    def test_first_commit_has_empty_diff(self, git_ready):
+        init_sha = git_ready.log()[-1].sha
+        result = git_ready.show_commit_diff(init_sha)
+        assert result is not None
+        _, diff = result
+        assert diff == ""
+
+    def test_returns_none_for_unknown(self, git_ready):
+        assert git_ready.show_commit_diff("deadbeef") is None
+
+
+class TestCommitInfoFormat:
+    def test_format_with_diff(self):
+        from nanobot.utils.gitstore import CommitInfo
+        c = CommitInfo(sha="abcd1234", message="test commit\nsecond line", timestamp="2026-04-02 12:00")
+        result = c.format(diff="some diff")
+        assert "test commit" in result
+        assert "`abcd1234`" in result
+        assert "some diff" in result
+
+    def test_format_without_diff(self):
+        from nanobot.utils.gitstore import CommitInfo
+        c = CommitInfo(sha="abcd1234", message="test", timestamp="2026-04-02 12:00")
+        result = c.format()
+        assert "(no file changes)" in result
+
+
+class TestRevert:
+    def test_returns_none_when_not_initialized(self, git):
+        assert git.revert("abc") is None
+
+    def test_undoes_commit_changes(self, git_ready):
+        """revert(sha) should undo the given commit by restoring to its parent."""
+        ws = git_ready._workspace
+        (ws / "SOUL.md").write_text("v2 content", encoding="utf-8")
+        git_ready.auto_commit("v2")
+
+        commits = git_ready.log()
+        # commits[0] = v2 (HEAD), commits[1] = init
+        # Revert v2 → restore to init's state (empty SOUL.md)
+        new_sha = git_ready.revert(commits[0].sha)
+        assert new_sha is not None
+        assert (ws / "SOUL.md").read_text(encoding="utf-8") == ""
+
+    def test_root_commit_returns_none(self, git_ready):
+        """Cannot revert the root commit (no parent to restore to)."""
+        commits = git_ready.log()
+        assert len(commits) == 1
+        assert git_ready.revert(commits[0].sha) is None
+
+    def test_invalid_sha_returns_none(self, git_ready):
+        assert git_ready.revert("deadbeef") is None
+
+
+class TestMemoryStoreGitProperty:
+    def test_git_property_exposes_gitstore(self, tmp_path):
+        from nanobot.agent.memory import MemoryStore
+        store = MemoryStore(tmp_path)
+        assert isinstance(store.git, GitStore)
+
+    def test_git_property_is_same_object(self, tmp_path):
+        from nanobot.agent.memory import MemoryStore
+        store = MemoryStore(tmp_path)
+        assert store.git is store._git

tests/agent/test_hook_composite.py

@@ -249,7 +249,8 @@ def _make_loop(tmp_path, hooks=None):
     with patch("nanobot.agent.loop.ContextBuilder"), \
          patch("nanobot.agent.loop.SessionManager"), \
          patch("nanobot.agent.loop.SubagentManager") as mock_sub_mgr, \
-         patch("nanobot.agent.loop.MemoryConsolidator"):
+         patch("nanobot.agent.loop.Consolidator"), \
+         patch("nanobot.agent.loop.Dream"):
         mock_sub_mgr.return_value.cancel_by_session = AsyncMock(return_value=0)
         loop = AgentLoop(
             bus=bus, provider=provider, workspace=tmp_path, hooks=hooks,

tests/agent/test_loop_consolidation_tokens.py

@@ -26,24 +26,24 @@ def _make_loop(tmp_path, *, estimated_tokens: int, context_window_tokens: int) -
         context_window_tokens=context_window_tokens,
     )
     loop.tools.get_definitions = MagicMock(return_value=[])
-    loop.memory_consolidator._SAFETY_BUFFER = 0
+    loop.consolidator._SAFETY_BUFFER = 0
     return loop
 
 
 @pytest.mark.asyncio
 async def test_prompt_below_threshold_does_not_consolidate(tmp_path) -> None:
     loop = _make_loop(tmp_path, estimated_tokens=100, context_window_tokens=200)
-    loop.memory_consolidator.consolidate_messages = AsyncMock(return_value=True)  # type: ignore[method-assign]
+    loop.consolidator.archive = AsyncMock(return_value=True)  # type: ignore[method-assign]
 
     await loop.process_direct("hello", session_key="cli:test")
 
-    loop.memory_consolidator.consolidate_messages.assert_not_awaited()
+    loop.consolidator.archive.assert_not_awaited()
 
 
 @pytest.mark.asyncio
 async def test_prompt_above_threshold_triggers_consolidation(tmp_path, monkeypatch) -> None:
     loop = _make_loop(tmp_path, estimated_tokens=1000, context_window_tokens=200)
-    loop.memory_consolidator.consolidate_messages = AsyncMock(return_value=True)  # type: ignore[method-assign]
+    loop.consolidator.archive = AsyncMock(return_value=True)  # type: ignore[method-assign]
     session = loop.sessions.get_or_create("cli:test")
     session.messages = [
         {"role": "user", "content": "u1", "timestamp": "2026-01-01T00:00:00"},
@@ -55,13 +55,13 @@ async def test_prompt_above_threshold_triggers_consolidation(tmp_path, monkeypat
 
     await loop.process_direct("hello", session_key="cli:test")
 
-    assert loop.memory_consolidator.consolidate_messages.await_count >= 1
+    assert loop.consolidator.archive.await_count >= 1
 
 
 @pytest.mark.asyncio
 async def test_prompt_above_threshold_archives_until_next_user_boundary(tmp_path, monkeypatch) -> None:
     loop = _make_loop(tmp_path, estimated_tokens=1000, context_window_tokens=200)
-    loop.memory_consolidator.consolidate_messages = AsyncMock(return_value=True)  # type: ignore[method-assign]
+    loop.consolidator.archive = AsyncMock(return_value=True)  # type: ignore[method-assign]
 
     session = loop.sessions.get_or_create("cli:test")
     session.messages = [
@@ -76,9 +76,9 @@ async def test_prompt_above_threshold_archives_until_next_user_boundary(tmp_path
     token_map = {"u1": 120, "a1": 120, "u2": 120, "a2": 120, "u3": 120}
     monkeypatch.setattr(memory_module, "estimate_message_tokens", lambda message: token_map[message["content"]])
 
-    await loop.memory_consolidator.maybe_consolidate_by_tokens(session)
+    await loop.consolidator.maybe_consolidate_by_tokens(session)
 
-    archived_chunk = loop.memory_consolidator.consolidate_messages.await_args.args[0]
+    archived_chunk = loop.consolidator.archive.await_args.args[0]
     assert [message["content"] for message in archived_chunk] == ["u1", "a1", "u2", "a2"]
     assert session.last_consolidated == 4
 
@@ -87,7 +87,7 @@ async def test_prompt_above_threshold_archives_until_next_user_boundary(tmp_path
 async def test_consolidation_loops_until_target_met(tmp_path, monkeypatch) -> None:
     """Verify maybe_consolidate_by_tokens keeps looping until under threshold."""
     loop = _make_loop(tmp_path, estimated_tokens=0, context_window_tokens=200)
-    loop.memory_consolidator.consolidate_messages = AsyncMock(return_value=True)  # type: ignore[method-assign]
+    loop.consolidator.archive = AsyncMock(return_value=True)  # type: ignore[method-assign]
 
     session = loop.sessions.get_or_create("cli:test")
     session.messages = [
@@ -110,12 +110,12 @@ async def test_consolidation_loops_until_target_met(tmp_path, monkeypatch) -> No
             return (300, "test")
         return (80, "test")
 
-    loop.memory_consolidator.estimate_session_prompt_tokens = mock_estimate  # type: ignore[method-assign]
+    loop.consolidator.estimate_session_prompt_tokens = mock_estimate  # type: ignore[method-assign]
     monkeypatch.setattr(memory_module, "estimate_message_tokens", lambda _m: 100)
 
-    await loop.memory_consolidator.maybe_consolidate_by_tokens(session)
+    await loop.consolidator.maybe_consolidate_by_tokens(session)
 
-    assert loop.memory_consolidator.consolidate_messages.await_count == 2
+    assert loop.consolidator.archive.await_count == 2
     assert session.last_consolidated == 6
 
 
@@ -123,7 +123,7 @@ async def test_consolidation_loops_until_target_met(tmp_path, monkeypatch) -> No
 async def test_consolidation_continues_below_trigger_until_half_target(tmp_path, monkeypatch) -> None:
     """Once triggered, consolidation should continue until it drops below half threshold."""
     loop = _make_loop(tmp_path, estimated_tokens=0, context_window_tokens=200)
-    loop.memory_consolidator.consolidate_messages = AsyncMock(return_value=True)  # type: ignore[method-assign]
+    loop.consolidator.archive = AsyncMock(return_value=True)  # type: ignore[method-assign]
 
     session = loop.sessions.get_or_create("cli:test")
     session.messages = [
@@ -147,12 +147,12 @@ async def test_consolidation_continues_below_trigger_until_half_target(tmp_path,
             return (150, "test")
         return (80, "test")
 
-    loop.memory_consolidator.estimate_session_prompt_tokens = mock_estimate  # type: ignore[method-assign]
+    loop.consolidator.estimate_session_prompt_tokens = mock_estimate  # type: ignore[method-assign]
     monkeypatch.setattr(memory_module, "estimate_message_tokens", lambda _m: 100)
 
-    await loop.memory_consolidator.maybe_consolidate_by_tokens(session)
+    await loop.consolidator.maybe_consolidate_by_tokens(session)
 
-    assert loop.memory_consolidator.consolidate_messages.await_count == 2
+    assert loop.consolidator.archive.await_count == 2
     assert session.last_consolidated == 6
 
 
@@ -166,7 +166,7 @@ async def test_preflight_consolidation_before_llm_call(tmp_path, monkeypatch) ->
     async def track_consolidate(messages):
         order.append("consolidate")
         return True
-    loop.memory_consolidator.consolidate_messages = track_consolidate  # type: ignore[method-assign]
+    loop.consolidator.archive = track_consolidate  # type: ignore[method-assign]
 
     async def track_llm(*args, **kwargs):
         order.append("llm")
@@ -187,7 +187,7 @@ async def test_preflight_consolidation_before_llm_call(tmp_path, monkeypatch) ->
     def mock_estimate(_session):
         call_count[0] += 1
         return (1000 if call_count[0] <= 1 else 80, "test")
-    loop.memory_consolidator.estimate_session_prompt_tokens = mock_estimate  # type: ignore[method-assign]
+    loop.consolidator.estimate_session_prompt_tokens = mock_estimate  # type: ignore[method-assign]
 
     await loop.process_direct("hello", session_key="cli:test")
 

tests/agent/test_memory_consolidation_types.py

@@ -1,478 +0,0 @@
-"""Test MemoryStore.consolidate() handles non-string tool call arguments.
-
-Regression test for https://github.com/HKUDS/nanobot/issues/1042
-When memory consolidation receives dict values instead of strings from the LLM
-tool call response, it should serialize them to JSON instead of raising TypeError.
-"""
-
-import json
-from pathlib import Path
-from unittest.mock import AsyncMock
-
-import pytest
-
-from nanobot.agent.memory import MemoryStore
-from nanobot.providers.base import LLMProvider, LLMResponse, ToolCallRequest
-
-
-def _make_messages(message_count: int = 30):
-    """Create a list of mock messages."""
-    return [
-        {"role": "user", "content": f"msg{i}", "timestamp": "2026-01-01 00:00"}
-        for i in range(message_count)
-    ]
-
-
-def _make_tool_response(history_entry, memory_update):
-    """Create an LLMResponse with a save_memory tool call."""
-    return LLMResponse(
-        content=None,
-        tool_calls=[
-            ToolCallRequest(
-                id="call_1",
-                name="save_memory",
-                arguments={
-                    "history_entry": history_entry,
-                    "memory_update": memory_update,
-                },
-            )
-        ],
-    )
-
-
-class ScriptedProvider(LLMProvider):
-    def __init__(self, responses: list[LLMResponse]):
-        super().__init__()
-        self._responses = list(responses)
-        self.calls = 0
-
-    async def chat(self, *args, **kwargs) -> LLMResponse:
-        self.calls += 1
-        if self._responses:
-            return self._responses.pop(0)
-        return LLMResponse(content="", tool_calls=[])
-
-    def get_default_model(self) -> str:
-        return "test-model"
-
-
-class TestMemoryConsolidationTypeHandling:
-    """Test that consolidation handles various argument types correctly."""
-
-    @pytest.mark.asyncio
-    async def test_string_arguments_work(self, tmp_path: Path) -> None:
-        """Normal case: LLM returns string arguments."""
-        store = MemoryStore(tmp_path)
-        provider = AsyncMock()
-        provider.chat = AsyncMock(
-            return_value=_make_tool_response(
-                history_entry="[2026-01-01] User discussed testing.",
-                memory_update="# Memory\nUser likes testing.",
-            )
-        )
-        provider.chat_with_retry = provider.chat
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is True
-        assert store.history_file.exists()
-        assert "[2026-01-01] User discussed testing." in store.history_file.read_text()
-        assert "User likes testing." in store.memory_file.read_text()
-
-    @pytest.mark.asyncio
-    async def test_dict_arguments_serialized_to_json(self, tmp_path: Path) -> None:
-        """Issue #1042: LLM returns dict instead of string — must not raise TypeError."""
-        store = MemoryStore(tmp_path)
-        provider = AsyncMock()
-        provider.chat = AsyncMock(
-            return_value=_make_tool_response(
-                history_entry={"timestamp": "2026-01-01", "summary": "User discussed testing."},
-                memory_update={"facts": ["User likes testing"], "topics": ["testing"]},
-            )
-        )
-        provider.chat_with_retry = provider.chat
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is True
-        assert store.history_file.exists()
-        history_content = store.history_file.read_text()
-        parsed = json.loads(history_content.strip())
-        assert parsed["summary"] == "User discussed testing."
-
-        memory_content = store.memory_file.read_text()
-        parsed_mem = json.loads(memory_content)
-        assert "User likes testing" in parsed_mem["facts"]
-
-    @pytest.mark.asyncio
-    async def test_string_arguments_as_raw_json(self, tmp_path: Path) -> None:
-        """Some providers return arguments as a JSON string instead of parsed dict."""
-        store = MemoryStore(tmp_path)
-        provider = AsyncMock()
-
-        response = LLMResponse(
-            content=None,
-            tool_calls=[
-                ToolCallRequest(
-                    id="call_1",
-                    name="save_memory",
-                    arguments=json.dumps({
-                        "history_entry": "[2026-01-01] User discussed testing.",
-                        "memory_update": "# Memory\nUser likes testing.",
-                    }),
-                )
-            ],
-        )
-        provider.chat = AsyncMock(return_value=response)
-        provider.chat_with_retry = provider.chat
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is True
-        assert "User discussed testing." in store.history_file.read_text()
-
-    @pytest.mark.asyncio
-    async def test_no_tool_call_returns_false(self, tmp_path: Path) -> None:
-        """When LLM doesn't use the save_memory tool, return False."""
-        store = MemoryStore(tmp_path)
-        provider = AsyncMock()
-        provider.chat = AsyncMock(
-            return_value=LLMResponse(content="I summarized the conversation.", tool_calls=[])
-        )
-        provider.chat_with_retry = provider.chat
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is False
-        assert not store.history_file.exists()
-
-    @pytest.mark.asyncio
-    async def test_skips_when_message_chunk_is_empty(self, tmp_path: Path) -> None:
-        """Consolidation should be a no-op when the selected chunk is empty."""
-        store = MemoryStore(tmp_path)
-        provider = AsyncMock()
-        provider.chat_with_retry = provider.chat
-        messages: list[dict] = []
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is True
-        provider.chat.assert_not_called()
-
-    @pytest.mark.asyncio
-    async def test_list_arguments_extracts_first_dict(self, tmp_path: Path) -> None:
-        """Some providers return arguments as a list - extract first element if it's a dict."""
-        store = MemoryStore(tmp_path)
-        provider = AsyncMock()
-
-        response = LLMResponse(
-            content=None,
-            tool_calls=[
-                ToolCallRequest(
-                    id="call_1",
-                    name="save_memory",
-                    arguments=[{
-                        "history_entry": "[2026-01-01] User discussed testing.",
-                        "memory_update": "# Memory\nUser likes testing.",
-                    }],
-                )
-            ],
-        )
-        provider.chat = AsyncMock(return_value=response)
-        provider.chat_with_retry = provider.chat
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is True
-        assert "User discussed testing." in store.history_file.read_text()
-        assert "User likes testing." in store.memory_file.read_text()
-
-    @pytest.mark.asyncio
-    async def test_list_arguments_empty_list_returns_false(self, tmp_path: Path) -> None:
-        """Empty list arguments should return False."""
-        store = MemoryStore(tmp_path)
-        provider = AsyncMock()
-
-        response = LLMResponse(
-            content=None,
-            tool_calls=[
-                ToolCallRequest(
-                    id="call_1",
-                    name="save_memory",
-                    arguments=[],
-                )
-            ],
-        )
-        provider.chat = AsyncMock(return_value=response)
-        provider.chat_with_retry = provider.chat
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is False
-
-    @pytest.mark.asyncio
-    async def test_list_arguments_non_dict_content_returns_false(self, tmp_path: Path) -> None:
-        """List with non-dict content should return False."""
-        store = MemoryStore(tmp_path)
-        provider = AsyncMock()
-
-        response = LLMResponse(
-            content=None,
-            tool_calls=[
-                ToolCallRequest(
-                    id="call_1",
-                    name="save_memory",
-                    arguments=["string", "content"],
-                )
-            ],
-        )
-        provider.chat = AsyncMock(return_value=response)
-        provider.chat_with_retry = provider.chat
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is False
-
-    @pytest.mark.asyncio
-    async def test_missing_history_entry_returns_false_without_writing(self, tmp_path: Path) -> None:
-        """Do not persist partial results when required fields are missing."""
-        store = MemoryStore(tmp_path)
-        provider = AsyncMock()
-        provider.chat_with_retry = AsyncMock(
-            return_value=LLMResponse(
-                content=None,
-                tool_calls=[
-                    ToolCallRequest(
-                        id="call_1",
-                        name="save_memory",
-                        arguments={"memory_update": "# Memory\nOnly memory update"},
-                    )
-                ],
-            )
-        )
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is False
-        assert not store.history_file.exists()
-        assert not store.memory_file.exists()
-
-    @pytest.mark.asyncio
-    async def test_missing_memory_update_returns_false_without_writing(self, tmp_path: Path) -> None:
-        """Do not append history if memory_update is missing."""
-        store = MemoryStore(tmp_path)
-        provider = AsyncMock()
-        provider.chat_with_retry = AsyncMock(
-            return_value=LLMResponse(
-                content=None,
-                tool_calls=[
-                    ToolCallRequest(
-                        id="call_1",
-                        name="save_memory",
-                        arguments={"history_entry": "[2026-01-01] Partial output."},
-                    )
-                ],
-            )
-        )
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is False
-        assert not store.history_file.exists()
-        assert not store.memory_file.exists()
-
-    @pytest.mark.asyncio
-    async def test_null_required_field_returns_false_without_writing(self, tmp_path: Path) -> None:
-        """Null required fields should be rejected before persistence."""
-        store = MemoryStore(tmp_path)
-        provider = AsyncMock()
-        provider.chat_with_retry = AsyncMock(
-            return_value=_make_tool_response(
-                history_entry=None,
-                memory_update="# Memory\nUser likes testing.",
-            )
-        )
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is False
-        assert not store.history_file.exists()
-        assert not store.memory_file.exists()
-
-    @pytest.mark.asyncio
-    async def test_empty_history_entry_returns_false_without_writing(self, tmp_path: Path) -> None:
-        """Empty history entries should be rejected to avoid blank archival records."""
-        store = MemoryStore(tmp_path)
-        provider = AsyncMock()
-        provider.chat_with_retry = AsyncMock(
-            return_value=_make_tool_response(
-                history_entry="   ",
-                memory_update="# Memory\nUser likes testing.",
-            )
-        )
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is False
-        assert not store.history_file.exists()
-        assert not store.memory_file.exists()
-
-    @pytest.mark.asyncio
-    async def test_retries_transient_error_then_succeeds(self, tmp_path: Path, monkeypatch) -> None:
-        store = MemoryStore(tmp_path)
-        provider = ScriptedProvider([
-            LLMResponse(content="503 server error", finish_reason="error"),
-            _make_tool_response(
-                history_entry="[2026-01-01] User discussed testing.",
-                memory_update="# Memory\nUser likes testing.",
-            ),
-        ])
-        messages = _make_messages(message_count=60)
-        delays: list[int] = []
-
-        async def _fake_sleep(delay: int) -> None:
-            delays.append(delay)
-
-        monkeypatch.setattr("nanobot.providers.base.asyncio.sleep", _fake_sleep)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is True
-        assert provider.calls == 2
-        assert delays == [1]
-
-    @pytest.mark.asyncio
-    async def test_consolidation_delegates_to_provider_defaults(self, tmp_path: Path) -> None:
-        """Consolidation no longer passes generation params — the provider owns them."""
-        store = MemoryStore(tmp_path)
-        provider = AsyncMock()
-        provider.chat_with_retry = AsyncMock(
-            return_value=_make_tool_response(
-                history_entry="[2026-01-01] User discussed testing.",
-                memory_update="# Memory\nUser likes testing.",
-            )
-        )
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is True
-        provider.chat_with_retry.assert_awaited_once()
-        _, kwargs = provider.chat_with_retry.await_args
-        assert kwargs["model"] == "test-model"
-        assert "temperature" not in kwargs
-        assert "max_tokens" not in kwargs
-        assert "reasoning_effort" not in kwargs
-
-    @pytest.mark.asyncio
-    async def test_tool_choice_fallback_on_unsupported_error(self, tmp_path: Path) -> None:
-        """Forced tool_choice rejected by provider -> retry with auto and succeed."""
-        store = MemoryStore(tmp_path)
-        error_resp = LLMResponse(
-            content="Error calling LLM: BadRequestError: "
-            "The tool_choice parameter does not support being set to required or object",
-            finish_reason="error",
-            tool_calls=[],
-        )
-        ok_resp = _make_tool_response(
-            history_entry="[2026-01-01] Fallback worked.",
-            memory_update="# Memory\nFallback OK.",
-        )
-
-        call_log: list[dict] = []
-
-        async def _tracking_chat(**kwargs):
-            call_log.append(kwargs)
-            return error_resp if len(call_log) == 1 else ok_resp
-
-        provider = AsyncMock()
-        provider.chat_with_retry = AsyncMock(side_effect=_tracking_chat)
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is True
-        assert len(call_log) == 2
-        assert isinstance(call_log[0]["tool_choice"], dict)
-        assert call_log[1]["tool_choice"] == "auto"
-        assert "Fallback worked." in store.history_file.read_text()
-
-    @pytest.mark.asyncio
-    async def test_tool_choice_fallback_auto_no_tool_call(self, tmp_path: Path) -> None:
-        """Forced rejected, auto retry also produces no tool call -> return False."""
-        store = MemoryStore(tmp_path)
-        error_resp = LLMResponse(
-            content="Error: tool_choice must be none or auto",
-            finish_reason="error",
-            tool_calls=[],
-        )
-        no_tool_resp = LLMResponse(
-            content="Here is a summary.",
-            finish_reason="stop",
-            tool_calls=[],
-        )
-
-        provider = AsyncMock()
-        provider.chat_with_retry = AsyncMock(side_effect=[error_resp, no_tool_resp])
-        messages = _make_messages(message_count=60)
-
-        result = await store.consolidate(messages, provider, "test-model")
-
-        assert result is False
-        assert not store.history_file.exists()
-
-    @pytest.mark.asyncio
-    async def test_raw_archive_after_consecutive_failures(self, tmp_path: Path) -> None:
-        """After 3 consecutive failures, raw-archive messages and return True."""
-        store = MemoryStore(tmp_path)
-        no_tool = LLMResponse(content="No tool call.", finish_reason="stop", tool_calls=[])
-        provider = AsyncMock()
-        provider.chat_with_retry = AsyncMock(return_value=no_tool)
-        messages = _make_messages(message_count=10)
-
-        assert await store.consolidate(messages, provider, "m") is False
-        assert await store.consolidate(messages, provider, "m") is False
-        assert await store.consolidate(messages, provider, "m") is True
-
-        assert store.history_file.exists()
-        content = store.history_file.read_text()
-        assert "[RAW]" in content
-        assert "10 messages" in content
-        assert "msg0" in content
-        assert not store.memory_file.exists()
-
-    @pytest.mark.asyncio
-    async def test_raw_archive_counter_resets_on_success(self, tmp_path: Path) -> None:
-        """A successful consolidation resets the failure counter."""
-        store = MemoryStore(tmp_path)
-        no_tool = LLMResponse(content="Nope.", finish_reason="stop", tool_calls=[])
-        ok_resp = _make_tool_response(
-            history_entry="[2026-01-01] OK.",
-            memory_update="# Memory\nOK.",
-        )
-        messages = _make_messages(message_count=10)
-
-        provider = AsyncMock()
-        provider.chat_with_retry = AsyncMock(return_value=no_tool)
-        assert await store.consolidate(messages, provider, "m") is False
-        assert await store.consolidate(messages, provider, "m") is False
-        assert store._consecutive_failures == 2
-
-        provider.chat_with_retry = AsyncMock(return_value=ok_resp)
-        assert await store.consolidate(messages, provider, "m") is True
-        assert store._consecutive_failures == 0
-
-        provider.chat_with_retry = AsyncMock(return_value=no_tool)
-        assert await store.consolidate(messages, provider, "m") is False
-        assert store._consecutive_failures == 1

tests/agent/test_memory_store.py

@@ -0,0 +1,267 @@
+"""Tests for the restructured MemoryStore — pure file I/O layer."""
+
+from datetime import datetime
+import json
+from pathlib import Path
+
+import pytest
+
+from nanobot.agent.memory import MemoryStore
+
+
+@pytest.fixture
+def store(tmp_path):
+    return MemoryStore(tmp_path)
+
+
+class TestMemoryStoreBasicIO:
+    def test_read_memory_returns_empty_when_missing(self, store):
+        assert store.read_memory() == ""
+
+    def test_write_and_read_memory(self, store):
+        store.write_memory("hello")
+        assert store.read_memory() == "hello"
+
+    def test_read_soul_returns_empty_when_missing(self, store):
+        assert store.read_soul() == ""
+
+    def test_write_and_read_soul(self, store):
+        store.write_soul("soul content")
+        assert store.read_soul() == "soul content"
+
+    def test_read_user_returns_empty_when_missing(self, store):
+        assert store.read_user() == ""
+
+    def test_write_and_read_user(self, store):
+        store.write_user("user content")
+        assert store.read_user() == "user content"
+
+    def test_get_memory_context_returns_empty_when_missing(self, store):
+        assert store.get_memory_context() == ""
+
+    def test_get_memory_context_returns_formatted_content(self, store):
+        store.write_memory("important fact")
+        ctx = store.get_memory_context()
+        assert "Long-term Memory" in ctx
+        assert "important fact" in ctx
+
+
+class TestHistoryWithCursor:
+    def test_append_history_returns_cursor(self, store):
+        cursor = store.append_history("event 1")
+        assert cursor == 1
+        cursor2 = store.append_history("event 2")
+        assert cursor2 == 2
+
+    def test_append_history_includes_cursor_in_file(self, store):
+        store.append_history("event 1")
+        content = store.read_file(store.history_file)
+        data = json.loads(content)
+        assert data["cursor"] == 1
+
+    def test_cursor_persists_across_appends(self, store):
+        store.append_history("event 1")
+        store.append_history("event 2")
+        cursor = store.append_history("event 3")
+        assert cursor == 3
+
+    def test_read_unprocessed_history(self, store):
+        store.append_history("event 1")
+        store.append_history("event 2")
+        store.append_history("event 3")
+        entries = store.read_unprocessed_history(since_cursor=1)
+        assert len(entries) == 2
+        assert entries[0]["cursor"] == 2
+
+    def test_read_unprocessed_history_returns_all_when_cursor_zero(self, store):
+        store.append_history("event 1")
+        store.append_history("event 2")
+        entries = store.read_unprocessed_history(since_cursor=0)
+        assert len(entries) == 2
+
+    def test_compact_history_drops_oldest(self, tmp_path):
+        store = MemoryStore(tmp_path, max_history_entries=2)
+        store.append_history("event 1")
+        store.append_history("event 2")
+        store.append_history("event 3")
+        store.append_history("event 4")
+        store.append_history("event 5")
+        store.compact_history()
+        entries = store.read_unprocessed_history(since_cursor=0)
+        assert len(entries) == 2
+        assert entries[0]["cursor"] in {4, 5}
+
+
+class TestDreamCursor:
+    def test_initial_cursor_is_zero(self, store):
+        assert store.get_last_dream_cursor() == 0
+
+    def test_set_and_get_cursor(self, store):
+        store.set_last_dream_cursor(5)
+        assert store.get_last_dream_cursor() == 5
+
+    def test_cursor_persists(self, store):
+        store.set_last_dream_cursor(3)
+        store2 = MemoryStore(store.workspace)
+        assert store2.get_last_dream_cursor() == 3
+
+
+class TestLegacyHistoryMigration:
+    def test_read_unprocessed_history_handles_entries_without_cursor(self, store):
+        """JSONL entries with cursor=1 are correctly parsed and returned."""
+        store.history_file.write_text(
+            '{"cursor": 1, "timestamp": "2026-03-30 14:30", "content": "Old event"}\n',
+            encoding="utf-8")
+        entries = store.read_unprocessed_history(since_cursor=0)
+        assert len(entries) == 1
+        assert entries[0]["cursor"] == 1
+
+    def test_migrates_legacy_history_md_preserving_partial_entries(self, tmp_path):
+        memory_dir = tmp_path / "memory"
+        memory_dir.mkdir()
+        legacy_file = memory_dir / "HISTORY.md"
+        legacy_content = (
+            "[2026-04-01 10:00] User prefers dark mode.\n\n"
+            "[2026-04-01 10:05] [RAW] 2 messages\n"
+            "[2026-04-01 10:04] USER: hello\n"
+            "[2026-04-01 10:04] ASSISTANT: hi\n\n"
+            "Legacy chunk without timestamp.\n"
+            "Keep whatever content we can recover.\n"
+        )
+        legacy_file.write_text(legacy_content, encoding="utf-8")
+
+        store = MemoryStore(tmp_path)
+        fallback_timestamp = datetime.fromtimestamp(
+            (memory_dir / "HISTORY.md.bak").stat().st_mtime,
+        ).strftime("%Y-%m-%d %H:%M")
+
+        entries = store.read_unprocessed_history(since_cursor=0)
+        assert [entry["cursor"] for entry in entries] == [1, 2, 3]
+        assert entries[0]["timestamp"] == "2026-04-01 10:00"
+        assert entries[0]["content"] == "User prefers dark mode."
+        assert entries[1]["timestamp"] == "2026-04-01 10:05"
+        assert entries[1]["content"].startswith("[RAW] 2 messages")
+        assert "USER: hello" in entries[1]["content"]
+        assert entries[2]["timestamp"] == fallback_timestamp
+        assert entries[2]["content"].startswith("Legacy chunk without timestamp.")
+        assert store.read_file(store._cursor_file).strip() == "3"
+        assert store.read_file(store._dream_cursor_file).strip() == "3"
+        assert not legacy_file.exists()
+        assert (memory_dir / "HISTORY.md.bak").read_text(encoding="utf-8") == legacy_content
+
+    def test_migrates_consecutive_entries_without_blank_lines(self, tmp_path):
+        memory_dir = tmp_path / "memory"
+        memory_dir.mkdir()
+        legacy_file = memory_dir / "HISTORY.md"
+        legacy_content = (
+            "[2026-04-01 10:00] First event.\n"
+            "[2026-04-01 10:01] Second event.\n"
+            "[2026-04-01 10:02] Third event.\n"
+        )
+        legacy_file.write_text(legacy_content, encoding="utf-8")
+
+        store = MemoryStore(tmp_path)
+
+        entries = store.read_unprocessed_history(since_cursor=0)
+        assert len(entries) == 3
+        assert [entry["content"] for entry in entries] == [
+            "First event.",
+            "Second event.",
+            "Third event.",
+        ]
+
+    def test_raw_archive_stays_single_entry_while_following_events_split(self, tmp_path):
+        memory_dir = tmp_path / "memory"
+        memory_dir.mkdir()
+        legacy_file = memory_dir / "HISTORY.md"
+        legacy_content = (
+            "[2026-04-01 10:05] [RAW] 2 messages\n"
+            "[2026-04-01 10:04] USER: hello\n"
+            "[2026-04-01 10:04] ASSISTANT: hi\n"
+            "[2026-04-01 10:06] Normal event after raw block.\n"
+        )
+        legacy_file.write_text(legacy_content, encoding="utf-8")
+
+        store = MemoryStore(tmp_path)
+
+        entries = store.read_unprocessed_history(since_cursor=0)
+        assert len(entries) == 2
+        assert entries[0]["content"].startswith("[RAW] 2 messages")
+        assert "USER: hello" in entries[0]["content"]
+        assert entries[1]["content"] == "Normal event after raw block."
+
+    def test_nonstandard_date_headers_still_start_new_entries(self, tmp_path):
+        memory_dir = tmp_path / "memory"
+        memory_dir.mkdir()
+        legacy_file = memory_dir / "HISTORY.md"
+        legacy_content = (
+            "[2026-03-25–2026-04-02] Multi-day summary.\n"
+            "[2026-03-26/27] Cross-day summary.\n"
+        )
+        legacy_file.write_text(legacy_content, encoding="utf-8")
+
+        store = MemoryStore(tmp_path)
+        fallback_timestamp = datetime.fromtimestamp(
+            (memory_dir / "HISTORY.md.bak").stat().st_mtime,
+        ).strftime("%Y-%m-%d %H:%M")
+
+        entries = store.read_unprocessed_history(since_cursor=0)
+        assert len(entries) == 2
+        assert entries[0]["timestamp"] == fallback_timestamp
+        assert entries[0]["content"] == "[2026-03-25–2026-04-02] Multi-day summary."
+        assert entries[1]["timestamp"] == fallback_timestamp
+        assert entries[1]["content"] == "[2026-03-26/27] Cross-day summary."
+
+    def test_existing_history_jsonl_skips_legacy_migration(self, tmp_path):
+        memory_dir = tmp_path / "memory"
+        memory_dir.mkdir()
+        history_file = memory_dir / "history.jsonl"
+        history_file.write_text(
+            '{"cursor": 7, "timestamp": "2026-04-01 12:00", "content": "existing"}\n',
+            encoding="utf-8",
+        )
+        legacy_file = memory_dir / "HISTORY.md"
+        legacy_file.write_text("[2026-04-01 10:00] legacy\n\n", encoding="utf-8")
+
+        store = MemoryStore(tmp_path)
+
+        entries = store.read_unprocessed_history(since_cursor=0)
+        assert len(entries) == 1
+        assert entries[0]["cursor"] == 7
+        assert entries[0]["content"] == "existing"
+        assert legacy_file.exists()
+        assert not (memory_dir / "HISTORY.md.bak").exists()
+
+    def test_empty_history_jsonl_still_allows_legacy_migration(self, tmp_path):
+        memory_dir = tmp_path / "memory"
+        memory_dir.mkdir()
+        history_file = memory_dir / "history.jsonl"
+        history_file.write_text("", encoding="utf-8")
+        legacy_file = memory_dir / "HISTORY.md"
+        legacy_file.write_text("[2026-04-01 10:00] legacy\n\n", encoding="utf-8")
+
+        store = MemoryStore(tmp_path)
+
+        entries = store.read_unprocessed_history(since_cursor=0)
+        assert len(entries) == 1
+        assert entries[0]["cursor"] == 1
+        assert entries[0]["timestamp"] == "2026-04-01 10:00"
+        assert entries[0]["content"] == "legacy"
+        assert not legacy_file.exists()
+        assert (memory_dir / "HISTORY.md.bak").exists()
+
+    def test_migrates_legacy_history_with_invalid_utf8_bytes(self, tmp_path):
+        memory_dir = tmp_path / "memory"
+        memory_dir.mkdir()
+        legacy_file = memory_dir / "HISTORY.md"
+        legacy_file.write_bytes(
+            b"[2026-04-01 10:00] Broken \xff data still needs migration.\n\n"
+        )
+
+        store = MemoryStore(tmp_path)
+
+        entries = store.read_unprocessed_history(since_cursor=0)
+        assert len(entries) == 1
+        assert entries[0]["timestamp"] == "2026-04-01 10:00"
+        assert "Broken" in entries[0]["content"]
+        assert "migration." in entries[0]["content"]

tests/agent/test_session_manager_history.py

@@ -173,6 +173,27 @@ def test_empty_session_history():
     assert history == []
 
 
+def test_get_history_preserves_reasoning_content():
+    session = Session(key="test:reasoning")
+    session.messages.append({"role": "user", "content": "hi"})
+    session.messages.append({
+        "role": "assistant",
+        "content": "done",
+        "reasoning_content": "hidden chain of thought",
+    })
+
+    history = session.get_history(max_messages=500)
+
+    assert history == [
+        {"role": "user", "content": "hi"},
+        {
+            "role": "assistant",
+            "content": "done",
+            "reasoning_content": "hidden chain of thought",
+        },
+    ]
+
+
 # --- Window cuts mid-group: assistant present but some tool results orphaned ---
 
 def test_window_cuts_mid_tool_group():

tests/agent/test_skills_loader.py

@@ -0,0 +1,252 @@
+"""Tests for nanobot.agent.skills.SkillsLoader."""
+
+from __future__ import annotations
+
+import json
+from pathlib import Path
+
+import pytest
+
+from nanobot.agent.skills import SkillsLoader
+
+
+def _write_skill(
+    base: Path,
+    name: str,
+    *,
+    metadata_json: dict | None = None,
+    body: str = "# Skill\n",
+) -> Path:
+    """Create ``base / name / SKILL.md`` with optional nanobot metadata JSON."""
+    skill_dir = base / name
+    skill_dir.mkdir(parents=True)
+    lines = ["---"]
+    if metadata_json is not None:
+        payload = json.dumps({"nanobot": metadata_json}, separators=(",", ":"))
+        lines.append(f'metadata: {payload}')
+    lines.extend(["---", "", body])
+    path = skill_dir / "SKILL.md"
+    path.write_text("\n".join(lines), encoding="utf-8")
+    return path
+
+
+def test_list_skills_empty_when_skills_dir_missing(tmp_path: Path) -> None:
+    workspace = tmp_path / "ws"
+    workspace.mkdir()
+    builtin = tmp_path / "builtin"
+    builtin.mkdir()
+    loader = SkillsLoader(workspace, builtin_skills_dir=builtin)
+    assert loader.list_skills(filter_unavailable=False) == []
+
+
+def test_list_skills_empty_when_skills_dir_exists_but_empty(tmp_path: Path) -> None:
+    workspace = tmp_path / "ws"
+    (workspace / "skills").mkdir(parents=True)
+    builtin = tmp_path / "builtin"
+    builtin.mkdir()
+    loader = SkillsLoader(workspace, builtin_skills_dir=builtin)
+    assert loader.list_skills(filter_unavailable=False) == []
+
+
+def test_list_skills_workspace_entry_shape_and_source(tmp_path: Path) -> None:
+    workspace = tmp_path / "ws"
+    skills_root = workspace / "skills"
+    skills_root.mkdir(parents=True)
+    skill_path = _write_skill(skills_root, "alpha", body="# Alpha")
+    builtin = tmp_path / "builtin"
+    builtin.mkdir()
+
+    loader = SkillsLoader(workspace, builtin_skills_dir=builtin)
+    entries = loader.list_skills(filter_unavailable=False)
+    assert entries == [
+        {"name": "alpha", "path": str(skill_path), "source": "workspace"},
+    ]
+
+
+def test_list_skills_skips_non_directories_and_missing_skill_md(tmp_path: Path) -> None:
+    workspace = tmp_path / "ws"
+    skills_root = workspace / "skills"
+    skills_root.mkdir(parents=True)
+    (skills_root / "not_a_dir.txt").write_text("x", encoding="utf-8")
+    (skills_root / "no_skill_md").mkdir()
+    ok_path = _write_skill(skills_root, "ok", body="# Ok")
+    builtin = tmp_path / "builtin"
+    builtin.mkdir()
+
+    loader = SkillsLoader(workspace, builtin_skills_dir=builtin)
+    entries = loader.list_skills(filter_unavailable=False)
+    names = {entry["name"] for entry in entries}
+    assert names == {"ok"}
+    assert entries[0]["path"] == str(ok_path)
+
+
+def test_list_skills_workspace_shadows_builtin_same_name(tmp_path: Path) -> None:
+    workspace = tmp_path / "ws"
+    ws_skills = workspace / "skills"
+    ws_skills.mkdir(parents=True)
+    ws_path = _write_skill(ws_skills, "dup", body="# Workspace wins")
+
+    builtin = tmp_path / "builtin"
+    _write_skill(builtin, "dup", body="# Builtin")
+
+    loader = SkillsLoader(workspace, builtin_skills_dir=builtin)
+    entries = loader.list_skills(filter_unavailable=False)
+    assert len(entries) == 1
+    assert entries[0]["source"] == "workspace"
+    assert entries[0]["path"] == str(ws_path)
+
+
+def test_list_skills_merges_workspace_and_builtin(tmp_path: Path) -> None:
+    workspace = tmp_path / "ws"
+    ws_skills = workspace / "skills"
+    ws_skills.mkdir(parents=True)
+    ws_path = _write_skill(ws_skills, "ws_only", body="# W")
+    builtin = tmp_path / "builtin"
+    bi_path = _write_skill(builtin, "bi_only", body="# B")
+
+    loader = SkillsLoader(workspace, builtin_skills_dir=builtin)
+    entries = sorted(loader.list_skills(filter_unavailable=False), key=lambda item: item["name"])
+    assert entries == [
+        {"name": "bi_only", "path": str(bi_path), "source": "builtin"},
+        {"name": "ws_only", "path": str(ws_path), "source": "workspace"},
+    ]
+
+
+def test_list_skills_builtin_omitted_when_dir_missing(tmp_path: Path) -> None:
+    workspace = tmp_path / "ws"
+    ws_skills = workspace / "skills"
+    ws_skills.mkdir(parents=True)
+    ws_path = _write_skill(ws_skills, "solo", body="# S")
+    missing_builtin = tmp_path / "no_such_builtin"
+
+    loader = SkillsLoader(workspace, builtin_skills_dir=missing_builtin)
+    entries = loader.list_skills(filter_unavailable=False)
+    assert entries == [{"name": "solo", "path": str(ws_path), "source": "workspace"}]
+
+
+def test_list_skills_filter_unavailable_excludes_unmet_bin_requirement(
+    tmp_path: Path, monkeypatch: pytest.MonkeyPatch
+) -> None:
+    workspace = tmp_path / "ws"
+    skills_root = workspace / "skills"
+    skills_root.mkdir(parents=True)
+    _write_skill(
+        skills_root,
+        "needs_bin",
+        metadata_json={"requires": {"bins": ["nanobot_test_fake_binary"]}},
+    )
+    builtin = tmp_path / "builtin"
+    builtin.mkdir()
+
+    def fake_which(cmd: str) -> str | None:
+        if cmd == "nanobot_test_fake_binary":
+            return None
+        return "/usr/bin/true"
+
+    monkeypatch.setattr("nanobot.agent.skills.shutil.which", fake_which)
+
+    loader = SkillsLoader(workspace, builtin_skills_dir=builtin)
+    assert loader.list_skills(filter_unavailable=True) == []
+
+
+def test_list_skills_filter_unavailable_includes_when_bin_requirement_met(
+    tmp_path: Path, monkeypatch: pytest.MonkeyPatch
+) -> None:
+    workspace = tmp_path / "ws"
+    skills_root = workspace / "skills"
+    skills_root.mkdir(parents=True)
+    skill_path = _write_skill(
+        skills_root,
+        "has_bin",
+        metadata_json={"requires": {"bins": ["nanobot_test_fake_binary"]}},
+    )
+    builtin = tmp_path / "builtin"
+    builtin.mkdir()
+
+    def fake_which(cmd: str) -> str | None:
+        if cmd == "nanobot_test_fake_binary":
+            return "/fake/nanobot_test_fake_binary"
+        return None
+
+    monkeypatch.setattr("nanobot.agent.skills.shutil.which", fake_which)
+
+    loader = SkillsLoader(workspace, builtin_skills_dir=builtin)
+    entries = loader.list_skills(filter_unavailable=True)
+    assert entries == [
+        {"name": "has_bin", "path": str(skill_path), "source": "workspace"},
+    ]
+
+
+def test_list_skills_filter_unavailable_false_keeps_unmet_requirements(
+    tmp_path: Path, monkeypatch: pytest.MonkeyPatch
+) -> None:
+    workspace = tmp_path / "ws"
+    skills_root = workspace / "skills"
+    skills_root.mkdir(parents=True)
+    skill_path = _write_skill(
+        skills_root,
+        "blocked",
+        metadata_json={"requires": {"bins": ["nanobot_test_fake_binary"]}},
+    )
+    builtin = tmp_path / "builtin"
+    builtin.mkdir()
+
+    monkeypatch.setattr("nanobot.agent.skills.shutil.which", lambda _cmd: None)
+
+    loader = SkillsLoader(workspace, builtin_skills_dir=builtin)
+    entries = loader.list_skills(filter_unavailable=False)
+    assert entries == [
+        {"name": "blocked", "path": str(skill_path), "source": "workspace"},
+    ]
+
+
+def test_list_skills_filter_unavailable_excludes_unmet_env_requirement(
+    tmp_path: Path, monkeypatch: pytest.MonkeyPatch
+) -> None:
+    workspace = tmp_path / "ws"
+    skills_root = workspace / "skills"
+    skills_root.mkdir(parents=True)
+    _write_skill(
+        skills_root,
+        "needs_env",
+        metadata_json={"requires": {"env": ["NANOBOT_SKILLS_TEST_ENV_VAR"]}},
+    )
+    builtin = tmp_path / "builtin"
+    builtin.mkdir()
+
+    monkeypatch.delenv("NANOBOT_SKILLS_TEST_ENV_VAR", raising=False)
+
+    loader = SkillsLoader(workspace, builtin_skills_dir=builtin)
+    assert loader.list_skills(filter_unavailable=True) == []
+
+
+def test_list_skills_openclaw_metadata_parsed_for_requirements(
+    tmp_path: Path, monkeypatch: pytest.MonkeyPatch
+) -> None:
+    workspace = tmp_path / "ws"
+    skills_root = workspace / "skills"
+    skills_root.mkdir(parents=True)
+    skill_dir = skills_root / "openclaw_skill"
+    skill_dir.mkdir(parents=True)
+    skill_path = skill_dir / "SKILL.md"
+    oc_payload = json.dumps({"openclaw": {"requires": {"bins": ["nanobot_oc_bin"]}}}, separators=(",", ":"))
+    skill_path.write_text(
+        "\n".join(["---", f"metadata: {oc_payload}", "---", "", "# OC"]),
+        encoding="utf-8",
+    )
+    builtin = tmp_path / "builtin"
+    builtin.mkdir()
+
+    monkeypatch.setattr("nanobot.agent.skills.shutil.which", lambda _cmd: None)
+
+    loader = SkillsLoader(workspace, builtin_skills_dir=builtin)
+    assert loader.list_skills(filter_unavailable=True) == []
+
+    monkeypatch.setattr(
+        "nanobot.agent.skills.shutil.which",
+        lambda cmd: "/x" if cmd == "nanobot_oc_bin" else None,
+    )
+    entries = loader.list_skills(filter_unavailable=True)
+    assert entries == [
+        {"name": "openclaw_skill", "path": str(skill_path), "source": "workspace"},
+    ]

tests/channels/test_email_channel.py

@@ -1,5 +1,6 @@
 from email.message import EmailMessage
 from datetime import date
+from pathlib import Path
 import imaplib
 
 import pytest
@@ -650,3 +651,224 @@ def test_check_authentication_results_method() -> None:
     spf, dkim = EmailChannel._check_authentication_results(parsed)
     assert spf is False
     assert dkim is True
+
+
+# ---------------------------------------------------------------------------
+# Attachment extraction tests
+# ---------------------------------------------------------------------------
+
+
+def _make_raw_email_with_attachment(
+    from_addr: str = "alice@example.com",
+    subject: str = "With attachment",
+    body: str = "See attached.",
+    attachment_name: str = "doc.pdf",
+    attachment_content: bytes = b"%PDF-1.4 fake pdf content",
+    attachment_mime: str = "application/pdf",
+    auth_results: str | None = None,
+) -> bytes:
+    msg = EmailMessage()
+    msg["From"] = from_addr
+    msg["To"] = "bot@example.com"
+    msg["Subject"] = subject
+    msg["Message-ID"] = "<m1@example.com>"
+    if auth_results:
+        msg["Authentication-Results"] = auth_results
+    msg.set_content(body)
+    maintype, subtype = attachment_mime.split("/", 1)
+    msg.add_attachment(
+        attachment_content,
+        maintype=maintype,
+        subtype=subtype,
+        filename=attachment_name,
+    )
+    return msg.as_bytes()
+
+
+def test_extract_attachments_saves_pdf(tmp_path, monkeypatch) -> None:
+    """PDF attachment is saved to media dir and path returned in media list."""
+    monkeypatch.setattr("nanobot.channels.email.get_media_dir", lambda ch: tmp_path)
+
+    raw = _make_raw_email_with_attachment()
+    fake = _make_fake_imap(raw)
+    monkeypatch.setattr("nanobot.channels.email.imaplib.IMAP4_SSL", lambda _h, _p: fake)
+
+    cfg = _make_config(allowed_attachment_types=["application/pdf"], verify_dkim=False, verify_spf=False)
+    channel = EmailChannel(cfg, MessageBus())
+    items = channel._fetch_new_messages()
+
+    assert len(items) == 1
+    assert len(items[0]["media"]) == 1
+    saved_path = Path(items[0]["media"][0])
+    assert saved_path.exists()
+    assert saved_path.read_bytes() == b"%PDF-1.4 fake pdf content"
+    assert "500_doc.pdf" in saved_path.name
+    assert "[attachment:" in items[0]["content"]
+
+
+def test_extract_attachments_disabled_by_default(monkeypatch) -> None:
+    """With no allowed_attachment_types (default), no attachments are extracted."""
+    raw = _make_raw_email_with_attachment()
+    fake = _make_fake_imap(raw)
+    monkeypatch.setattr("nanobot.channels.email.imaplib.IMAP4_SSL", lambda _h, _p: fake)
+
+    cfg = _make_config(verify_dkim=False, verify_spf=False)
+    assert cfg.allowed_attachment_types == []
+    channel = EmailChannel(cfg, MessageBus())
+    items = channel._fetch_new_messages()
+
+    assert len(items) == 1
+    assert items[0]["media"] == []
+    assert "[attachment:" not in items[0]["content"]
+
+
+def test_extract_attachments_mime_type_filter(tmp_path, monkeypatch) -> None:
+    """Non-allowed MIME types are skipped."""
+    monkeypatch.setattr("nanobot.channels.email.get_media_dir", lambda ch: tmp_path)
+
+    raw = _make_raw_email_with_attachment(
+        attachment_name="image.png",
+        attachment_content=b"\x89PNG fake",
+        attachment_mime="image/png",
+    )
+    fake = _make_fake_imap(raw)
+    monkeypatch.setattr("nanobot.channels.email.imaplib.IMAP4_SSL", lambda _h, _p: fake)
+
+    cfg = _make_config(
+        allowed_attachment_types=["application/pdf"],
+        verify_dkim=False,
+        verify_spf=False,
+    )
+    channel = EmailChannel(cfg, MessageBus())
+    items = channel._fetch_new_messages()
+
+    assert len(items) == 1
+    assert items[0]["media"] == []
+
+
+def test_extract_attachments_empty_allowed_types_rejects_all(tmp_path, monkeypatch) -> None:
+    """Empty allowed_attachment_types means no types are accepted."""
+    monkeypatch.setattr("nanobot.channels.email.get_media_dir", lambda ch: tmp_path)
+
+    raw = _make_raw_email_with_attachment(
+        attachment_name="image.png",
+        attachment_content=b"\x89PNG fake",
+        attachment_mime="image/png",
+    )
+    fake = _make_fake_imap(raw)
+    monkeypatch.setattr("nanobot.channels.email.imaplib.IMAP4_SSL", lambda _h, _p: fake)
+
+    cfg = _make_config(
+        allowed_attachment_types=[],
+        verify_dkim=False,
+        verify_spf=False,
+    )
+    channel = EmailChannel(cfg, MessageBus())
+    items = channel._fetch_new_messages()
+
+    assert len(items) == 1
+    assert items[0]["media"] == []
+
+
+def test_extract_attachments_wildcard_pattern(tmp_path, monkeypatch) -> None:
+    """Glob patterns like 'image/*' match attachment MIME types."""
+    monkeypatch.setattr("nanobot.channels.email.get_media_dir", lambda ch: tmp_path)
+
+    raw = _make_raw_email_with_attachment(
+        attachment_name="photo.jpg",
+        attachment_content=b"\xff\xd8\xff fake jpeg",
+        attachment_mime="image/jpeg",
+    )
+    fake = _make_fake_imap(raw)
+    monkeypatch.setattr("nanobot.channels.email.imaplib.IMAP4_SSL", lambda _h, _p: fake)
+
+    cfg = _make_config(
+        allowed_attachment_types=["image/*"],
+        verify_dkim=False,
+        verify_spf=False,
+    )
+    channel = EmailChannel(cfg, MessageBus())
+    items = channel._fetch_new_messages()
+
+    assert len(items) == 1
+    assert len(items[0]["media"]) == 1
+
+
+def test_extract_attachments_size_limit(tmp_path, monkeypatch) -> None:
+    """Attachments exceeding max_attachment_size are skipped."""
+    monkeypatch.setattr("nanobot.channels.email.get_media_dir", lambda ch: tmp_path)
+
+    raw = _make_raw_email_with_attachment(
+        attachment_content=b"x" * 1000,
+    )
+    fake = _make_fake_imap(raw)
+    monkeypatch.setattr("nanobot.channels.email.imaplib.IMAP4_SSL", lambda _h, _p: fake)
+
+    cfg = _make_config(
+        allowed_attachment_types=["*"],
+        max_attachment_size=500,
+        verify_dkim=False,
+        verify_spf=False,
+    )
+    channel = EmailChannel(cfg, MessageBus())
+    items = channel._fetch_new_messages()
+
+    assert len(items) == 1
+    assert items[0]["media"] == []
+
+
+def test_extract_attachments_max_count(tmp_path, monkeypatch) -> None:
+    """Only max_attachments_per_email are saved."""
+    monkeypatch.setattr("nanobot.channels.email.get_media_dir", lambda ch: tmp_path)
+
+    # Build email with 3 attachments
+    msg = EmailMessage()
+    msg["From"] = "alice@example.com"
+    msg["To"] = "bot@example.com"
+    msg["Subject"] = "Many attachments"
+    msg["Message-ID"] = "<m1@example.com>"
+    msg.set_content("See attached.")
+    for i in range(3):
+        msg.add_attachment(
+            f"content {i}".encode(),
+            maintype="application",
+            subtype="pdf",
+            filename=f"doc{i}.pdf",
+        )
+    raw = msg.as_bytes()
+
+    fake = _make_fake_imap(raw)
+    monkeypatch.setattr("nanobot.channels.email.imaplib.IMAP4_SSL", lambda _h, _p: fake)
+
+    cfg = _make_config(
+        allowed_attachment_types=["*"],
+        max_attachments_per_email=2,
+        verify_dkim=False,
+        verify_spf=False,
+    )
+    channel = EmailChannel(cfg, MessageBus())
+    items = channel._fetch_new_messages()
+
+    assert len(items) == 1
+    assert len(items[0]["media"]) == 2
+
+
+def test_extract_attachments_sanitizes_filename(tmp_path, monkeypatch) -> None:
+    """Path traversal in filenames is neutralized."""
+    monkeypatch.setattr("nanobot.channels.email.get_media_dir", lambda ch: tmp_path)
+
+    raw = _make_raw_email_with_attachment(
+        attachment_name="../../../etc/passwd",
+    )
+    fake = _make_fake_imap(raw)
+    monkeypatch.setattr("nanobot.channels.email.imaplib.IMAP4_SSL", lambda _h, _p: fake)
+
+    cfg = _make_config(allowed_attachment_types=["*"], verify_dkim=False, verify_spf=False)
+    channel = EmailChannel(cfg, MessageBus())
+    items = channel._fetch_new_messages()
+
+    assert len(items) == 1
+    assert len(items[0]["media"]) == 1
+    saved_path = Path(items[0]["media"][0])
+    # File must be inside the media dir, not escaped via path traversal
+    assert saved_path.parent == tmp_path

tests/channels/test_feishu_mention.py

@@ -0,0 +1,62 @@
+"""Tests for Feishu _is_bot_mentioned logic."""
+
+from types import SimpleNamespace
+
+import pytest
+
+from nanobot.channels.feishu import FeishuChannel
+
+
+def _make_channel(bot_open_id: str | None = None) -> FeishuChannel:
+    config = SimpleNamespace(
+        app_id="test_id",
+        app_secret="test_secret",
+        verification_token="",
+        event_encrypt_key="",
+        group_policy="mention",
+    )
+    ch = FeishuChannel.__new__(FeishuChannel)
+    ch.config = config
+    ch._bot_open_id = bot_open_id
+    return ch
+
+
+def _make_message(mentions=None, content="hello"):
+    return SimpleNamespace(content=content, mentions=mentions)
+
+
+def _make_mention(open_id: str, user_id: str | None = None):
+    mid = SimpleNamespace(open_id=open_id, user_id=user_id)
+    return SimpleNamespace(id=mid)
+
+
+class TestIsBotMentioned:
+    def test_exact_match_with_bot_open_id(self):
+        ch = _make_channel(bot_open_id="ou_bot123")
+        msg = _make_message(mentions=[_make_mention("ou_bot123")])
+        assert ch._is_bot_mentioned(msg) is True
+
+    def test_no_match_different_bot(self):
+        ch = _make_channel(bot_open_id="ou_bot123")
+        msg = _make_message(mentions=[_make_mention("ou_other_bot")])
+        assert ch._is_bot_mentioned(msg) is False
+
+    def test_at_all_always_matches(self):
+        ch = _make_channel(bot_open_id="ou_bot123")
+        msg = _make_message(content="@_all hello")
+        assert ch._is_bot_mentioned(msg) is True
+
+    def test_fallback_heuristic_when_no_bot_open_id(self):
+        ch = _make_channel(bot_open_id=None)
+        msg = _make_message(mentions=[_make_mention("ou_some_bot", user_id=None)])
+        assert ch._is_bot_mentioned(msg) is True
+
+    def test_fallback_ignores_user_mentions(self):
+        ch = _make_channel(bot_open_id=None)
+        msg = _make_message(mentions=[_make_mention("ou_user", user_id="u_12345")])
+        assert ch._is_bot_mentioned(msg) is False
+
+    def test_no_mentions_returns_false(self):
+        ch = _make_channel(bot_open_id="ou_bot123")
+        msg = _make_message(mentions=None)
+        assert ch._is_bot_mentioned(msg) is False

tests/channels/test_feishu_reaction.py

@@ -0,0 +1,238 @@
+"""Tests for Feishu reaction add/remove and auto-cleanup on stream end."""
+from types import SimpleNamespace
+from unittest.mock import AsyncMock, MagicMock, patch
+
+import pytest
+
+from nanobot.bus.queue import MessageBus
+from nanobot.channels.feishu import FeishuChannel, FeishuConfig, _FeishuStreamBuf
+
+
+def _make_channel() -> FeishuChannel:
+    config = FeishuConfig(
+        enabled=True,
+        app_id="cli_test",
+        app_secret="secret",
+        allow_from=["*"],
+    )
+    ch = FeishuChannel(config, MessageBus())
+    ch._client = MagicMock()
+    ch._loop = None
+    return ch
+
+
+def _mock_reaction_create_response(reaction_id: str = "reaction_001", success: bool = True):
+    resp = MagicMock()
+    resp.success.return_value = success
+    resp.code = 0 if success else 99999
+    resp.msg = "ok" if success else "error"
+    if success:
+        resp.data = SimpleNamespace(reaction_id=reaction_id)
+    else:
+        resp.data = None
+    return resp
+
+
+# ── _add_reaction_sync ──────────────────────────────────────────────────────
+
+
+class TestAddReactionSync:
+    def test_returns_reaction_id_on_success(self):
+        ch = _make_channel()
+        ch._client.im.v1.message_reaction.create.return_value = _mock_reaction_create_response("rx_42")
+        result = ch._add_reaction_sync("om_001", "THUMBSUP")
+        assert result == "rx_42"
+
+    def test_returns_none_when_response_fails(self):
+        ch = _make_channel()
+        ch._client.im.v1.message_reaction.create.return_value = _mock_reaction_create_response(success=False)
+        assert ch._add_reaction_sync("om_001", "THUMBSUP") is None
+
+    def test_returns_none_when_response_data_is_none(self):
+        ch = _make_channel()
+        resp = MagicMock()
+        resp.success.return_value = True
+        resp.data = None
+        ch._client.im.v1.message_reaction.create.return_value = resp
+        assert ch._add_reaction_sync("om_001", "THUMBSUP") is None
+
+    def test_returns_none_on_exception(self):
+        ch = _make_channel()
+        ch._client.im.v1.message_reaction.create.side_effect = RuntimeError("network error")
+        assert ch._add_reaction_sync("om_001", "THUMBSUP") is None
+
+
+# ── _add_reaction (async) ───────────────────────────────────────────────────
+
+
+class TestAddReactionAsync:
+    @pytest.mark.asyncio
+    async def test_returns_reaction_id(self):
+        ch = _make_channel()
+        ch._add_reaction_sync = MagicMock(return_value="rx_99")
+        result = await ch._add_reaction("om_001", "EYES")
+        assert result == "rx_99"
+
+    @pytest.mark.asyncio
+    async def test_returns_none_when_no_client(self):
+        ch = _make_channel()
+        ch._client = None
+        result = await ch._add_reaction("om_001", "THUMBSUP")
+        assert result is None
+
+
+# ── _remove_reaction_sync ───────────────────────────────────────────────────
+
+
+class TestRemoveReactionSync:
+    def test_calls_delete_on_success(self):
+        ch = _make_channel()
+        resp = MagicMock()
+        resp.success.return_value = True
+        ch._client.im.v1.message_reaction.delete.return_value = resp
+
+        ch._remove_reaction_sync("om_001", "rx_42")
+
+        ch._client.im.v1.message_reaction.delete.assert_called_once()
+
+    def test_handles_failure_gracefully(self):
+        ch = _make_channel()
+        resp = MagicMock()
+        resp.success.return_value = False
+        resp.code = 99999
+        resp.msg = "not found"
+        ch._client.im.v1.message_reaction.delete.return_value = resp
+
+        # Should not raise
+        ch._remove_reaction_sync("om_001", "rx_42")
+
+    def test_handles_exception_gracefully(self):
+        ch = _make_channel()
+        ch._client.im.v1.message_reaction.delete.side_effect = RuntimeError("network error")
+
+        # Should not raise
+        ch._remove_reaction_sync("om_001", "rx_42")
+
+
+# ── _remove_reaction (async) ────────────────────────────────────────────────
+
+
+class TestRemoveReactionAsync:
+    @pytest.mark.asyncio
+    async def test_calls_sync_helper(self):
+        ch = _make_channel()
+        ch._remove_reaction_sync = MagicMock()
+
+        await ch._remove_reaction("om_001", "rx_42")
+
+        ch._remove_reaction_sync.assert_called_once_with("om_001", "rx_42")
+
+    @pytest.mark.asyncio
+    async def test_noop_when_no_client(self):
+        ch = _make_channel()
+        ch._client = None
+        ch._remove_reaction_sync = MagicMock()
+
+        await ch._remove_reaction("om_001", "rx_42")
+
+        ch._remove_reaction_sync.assert_not_called()
+
+    @pytest.mark.asyncio
+    async def test_noop_when_reaction_id_is_empty(self):
+        ch = _make_channel()
+        ch._remove_reaction_sync = MagicMock()
+
+        await ch._remove_reaction("om_001", "")
+
+        ch._remove_reaction_sync.assert_not_called()
+
+    @pytest.mark.asyncio
+    async def test_noop_when_reaction_id_is_none(self):
+        ch = _make_channel()
+        ch._remove_reaction_sync = MagicMock()
+
+        await ch._remove_reaction("om_001", None)
+
+        ch._remove_reaction_sync.assert_not_called()
+
+
+# ── send_delta stream end: reaction auto-cleanup ────────────────────────────
+
+
+class TestStreamEndReactionCleanup:
+    @pytest.mark.asyncio
+    async def test_removes_reaction_on_stream_end(self):
+        ch = _make_channel()
+        ch._stream_bufs["oc_chat1"] = _FeishuStreamBuf(
+            text="Done", card_id="card_1", sequence=3, last_edit=0.0,
+        )
+        ch._client.cardkit.v1.card_element.content.return_value = MagicMock(success=MagicMock(return_value=True))
+        ch._client.cardkit.v1.card.settings.return_value = MagicMock(success=MagicMock(return_value=True))
+        ch._remove_reaction = AsyncMock()
+
+        await ch.send_delta(
+            "oc_chat1", "",
+            metadata={"_stream_end": True, "message_id": "om_001", "reaction_id": "rx_42"},
+        )
+
+        ch._remove_reaction.assert_called_once_with("om_001", "rx_42")
+
+    @pytest.mark.asyncio
+    async def test_no_removal_when_message_id_missing(self):
+        ch = _make_channel()
+        ch._stream_bufs["oc_chat1"] = _FeishuStreamBuf(
+            text="Done", card_id="card_1", sequence=3, last_edit=0.0,
+        )
+        ch._client.cardkit.v1.card_element.content.return_value = MagicMock(success=MagicMock(return_value=True))
+        ch._client.cardkit.v1.card.settings.return_value = MagicMock(success=MagicMock(return_value=True))
+        ch._remove_reaction = AsyncMock()
+
+        await ch.send_delta(
+            "oc_chat1", "",
+            metadata={"_stream_end": True, "reaction_id": "rx_42"},
+        )
+
+        ch._remove_reaction.assert_not_called()
+
+    @pytest.mark.asyncio
+    async def test_no_removal_when_reaction_id_missing(self):
+        ch = _make_channel()
+        ch._stream_bufs["oc_chat1"] = _FeishuStreamBuf(
+            text="Done", card_id="card_1", sequence=3, last_edit=0.0,
+        )
+        ch._client.cardkit.v1.card_element.content.return_value = MagicMock(success=MagicMock(return_value=True))
+        ch._client.cardkit.v1.card.settings.return_value = MagicMock(success=MagicMock(return_value=True))
+        ch._remove_reaction = AsyncMock()
+
+        await ch.send_delta(
+            "oc_chat1", "",
+            metadata={"_stream_end": True, "message_id": "om_001"},
+        )
+
+        ch._remove_reaction.assert_not_called()
+
+    @pytest.mark.asyncio
+    async def test_no_removal_when_both_ids_missing(self):
+        ch = _make_channel()
+        ch._stream_bufs["oc_chat1"] = _FeishuStreamBuf(
+            text="Done", card_id="card_1", sequence=3, last_edit=0.0,
+        )
+        ch._client.cardkit.v1.card_element.content.return_value = MagicMock(success=MagicMock(return_value=True))
+        ch._client.cardkit.v1.card.settings.return_value = MagicMock(success=MagicMock(return_value=True))
+        ch._remove_reaction = AsyncMock()
+
+        await ch.send_delta("oc_chat1", "", metadata={"_stream_end": True})
+
+        ch._remove_reaction.assert_not_called()
+
+    @pytest.mark.asyncio
+    async def test_no_removal_when_not_stream_end(self):
+        ch = _make_channel()
+        ch._remove_reaction = AsyncMock()
+
+        await ch.send_delta(
+            "oc_chat1", "more text",
+            metadata={"message_id": "om_001", "reaction_id": "rx_42"},
+        )
+
+        ch._remove_reaction.assert_not_called()

tests/channels/test_telegram_channel.py

@@ -32,8 +32,10 @@ class _FakeHTTPXRequest:
 class _FakeUpdater:
     def __init__(self, on_start_polling) -> None:
         self._on_start_polling = on_start_polling
+        self.start_polling_kwargs = None
 
     async def start_polling(self, **kwargs) -> None:
+        self.start_polling_kwargs = kwargs
         self._on_start_polling()
 
 
@@ -184,7 +186,11 @@ async def test_start_creates_separate_pools_with_proxy(monkeypatch) -> None:
     assert poll_req.kwargs["connection_pool_size"] == 4
     assert builder.request_value is api_req
     assert builder.get_updates_request_value is poll_req
+    assert callable(app.updater.start_polling_kwargs["error_callback"])
     assert any(cmd.command == "status" for cmd in app.bot.commands)
+    assert any(cmd.command == "dream" for cmd in app.bot.commands)
+    assert any(cmd.command == "dream_log" for cmd in app.bot.commands)
+    assert any(cmd.command == "dream_restore" for cmd in app.bot.commands)
 
 
 @pytest.mark.asyncio
@@ -304,6 +310,26 @@ async def test_on_error_logs_network_issues_as_warning(monkeypatch) -> None:
     assert recorded == [("warning", "Telegram network issue: proxy disconnected")]
 
 
+@pytest.mark.asyncio
+async def test_on_error_summarizes_empty_network_error(monkeypatch) -> None:
+    from telegram.error import NetworkError
+
+    channel = TelegramChannel(
+        TelegramConfig(enabled=True, token="123:abc", allow_from=["*"]),
+        MessageBus(),
+    )
+    recorded: list[tuple[str, str]] = []
+
+    monkeypatch.setattr(
+        "nanobot.channels.telegram.logger.warning",
+        lambda message, error: recorded.append(("warning", message.format(error))),
+    )
+
+    await channel._on_error(object(), SimpleNamespace(error=NetworkError("")))
+
+    assert recorded == [("warning", "Telegram network issue: NetworkError")]
+
+
 @pytest.mark.asyncio
 async def test_on_error_keeps_non_network_exceptions_as_error(monkeypatch) -> None:
     channel = TelegramChannel(
@@ -359,6 +385,32 @@ async def test_send_delta_stream_end_treats_not_modified_as_success() -> None:
     assert "123" not in channel._stream_bufs
 
 
+@pytest.mark.asyncio
+async def test_send_delta_stream_end_splits_oversized_reply() -> None:
+    """Final streamed reply exceeding Telegram limit is split into chunks."""
+    from nanobot.channels.telegram import TELEGRAM_MAX_MESSAGE_LEN
+
+    channel = TelegramChannel(
+        TelegramConfig(enabled=True, token="123:abc", allow_from=["*"]),
+        MessageBus(),
+    )
+    channel._app = _FakeApp(lambda: None)
+    channel._app.bot.edit_message_text = AsyncMock()
+    channel._app.bot.send_message = AsyncMock(return_value=SimpleNamespace(message_id=99))
+
+    oversized = "x" * (TELEGRAM_MAX_MESSAGE_LEN + 500)
+    channel._stream_bufs["123"] = _StreamBuf(text=oversized, message_id=7, last_edit=0.0)
+
+    await channel.send_delta("123", "", {"_stream_end": True})
+
+    channel._app.bot.edit_message_text.assert_called_once()
+    edit_text = channel._app.bot.edit_message_text.call_args.kwargs.get("text", "")
+    assert len(edit_text) <= TELEGRAM_MAX_MESSAGE_LEN
+
+    channel._app.bot.send_message.assert_called_once()
+    assert "123" not in channel._stream_bufs
+
+
 @pytest.mark.asyncio
 async def test_send_delta_new_stream_id_replaces_stale_buffer() -> None:
     channel = TelegramChannel(
@@ -398,6 +450,23 @@ async def test_send_delta_incremental_edit_treats_not_modified_as_success() -> N
     assert channel._stream_bufs["123"].last_edit > 0.0
 
 
+@pytest.mark.asyncio
+async def test_send_delta_initial_send_keeps_message_in_thread() -> None:
+    channel = TelegramChannel(
+        TelegramConfig(enabled=True, token="123:abc", allow_from=["*"]),
+        MessageBus(),
+    )
+    channel._app = _FakeApp(lambda: None)
+
+    await channel.send_delta(
+        "123",
+        "hello",
+        {"_stream_delta": True, "_stream_id": "s:0", "message_thread_id": 42},
+    )
+
+    assert channel._app.bot.sent_messages[0]["message_thread_id"] == 42
+
+
 def test_derive_topic_session_key_uses_thread_id() -> None:
     message = SimpleNamespace(
         chat=SimpleNamespace(type="supergroup"),
@@ -408,6 +477,27 @@ def test_derive_topic_session_key_uses_thread_id() -> None:
     assert TelegramChannel._derive_topic_session_key(message) == "telegram:-100123:topic:42"
 
 
+def test_derive_topic_session_key_private_dm_thread() -> None:
+    """Private DM threads (Telegram Threaded Mode) must get their own session key."""
+    message = SimpleNamespace(
+        chat=SimpleNamespace(type="private"),
+        chat_id=999,
+        message_thread_id=7,
+    )
+    assert TelegramChannel._derive_topic_session_key(message) == "telegram:999:topic:7"
+
+
+def test_derive_topic_session_key_none_without_thread() -> None:
+    """No thread id → no topic session key, regardless of chat type."""
+    for chat_type in ("private", "supergroup", "group"):
+        message = SimpleNamespace(
+            chat=SimpleNamespace(type=chat_type),
+            chat_id=123,
+            message_thread_id=None,
+        )
+        assert TelegramChannel._derive_topic_session_key(message) is None
+
+
 def test_get_extension_falls_back_to_original_filename() -> None:
     channel = TelegramChannel(TelegramConfig(), MessageBus())
 
@@ -962,6 +1052,48 @@ async def test_forward_command_does_not_inject_reply_context() -> None:
     assert handled[0]["content"] == "/new"
 
 
+@pytest.mark.asyncio
+async def test_forward_command_preserves_dream_log_args_and_strips_bot_suffix() -> None:
+    channel = TelegramChannel(
+        TelegramConfig(enabled=True, token="123:abc", allow_from=["*"], group_policy="open"),
+        MessageBus(),
+    )
+    channel._app = _FakeApp(lambda: None)
+    handled = []
+
+    async def capture_handle(**kwargs) -> None:
+        handled.append(kwargs)
+
+    channel._handle_message = capture_handle
+    update = _make_telegram_update(text="/dream-log@nanobot_test deadbeef", reply_to_message=None)
+
+    await channel._forward_command(update, None)
+
+    assert len(handled) == 1
+    assert handled[0]["content"] == "/dream-log deadbeef"
+
+
+@pytest.mark.asyncio
+async def test_forward_command_normalizes_telegram_safe_dream_aliases() -> None:
+    channel = TelegramChannel(
+        TelegramConfig(enabled=True, token="123:abc", allow_from=["*"], group_policy="open"),
+        MessageBus(),
+    )
+    channel._app = _FakeApp(lambda: None)
+    handled = []
+
+    async def capture_handle(**kwargs) -> None:
+        handled.append(kwargs)
+
+    channel._handle_message = capture_handle
+    update = _make_telegram_update(text="/dream_restore@nanobot_test deadbeef", reply_to_message=None)
+
+    await channel._forward_command(update, None)
+
+    assert len(handled) == 1
+    assert handled[0]["content"] == "/dream-restore deadbeef"
+
+
 @pytest.mark.asyncio
 async def test_on_help_includes_restart_command() -> None:
     channel = TelegramChannel(
@@ -977,3 +1109,6 @@ async def test_on_help_includes_restart_command() -> None:
     help_text = update.message.reply_text.await_args.args[0]
     assert "/restart" in help_text
     assert "/status" in help_text
+    assert "/dream" in help_text
+    assert "/dream-log" in help_text
+    assert "/dream-restore" in help_text

tests/channels/test_whatsapp_channel.py

@@ -1,12 +1,18 @@
 """Tests for WhatsApp channel outbound media support."""
 
 import json
+import os
+import sys
+import types
 from unittest.mock import AsyncMock, MagicMock
 
 import pytest
 
 from nanobot.bus.events import OutboundMessage
-from nanobot.channels.whatsapp import WhatsAppChannel
+from nanobot.channels.whatsapp import (
+    WhatsAppChannel,
+    _load_or_create_bridge_token,
+)
 
 
 def _make_channel() -> WhatsAppChannel:
@@ -155,3 +161,197 @@ async def test_group_policy_mention_accepts_mentioned_group_message():
     kwargs = ch._handle_message.await_args.kwargs
     assert kwargs["chat_id"] == "12345@g.us"
     assert kwargs["sender_id"] == "user"
+
+
+@pytest.mark.asyncio
+async def test_sender_id_prefers_phone_jid_over_lid():
+    """sender_id should resolve to phone number when @s.whatsapp.net JID is present."""
+    ch = WhatsAppChannel({"enabled": True}, MagicMock())
+    ch._handle_message = AsyncMock()
+
+    await ch._handle_bridge_message(
+        json.dumps({
+            "type": "message",
+            "id": "lid1",
+            "sender": "ABC123@lid.whatsapp.net",
+            "pn": "5551234@s.whatsapp.net",
+            "content": "hi",
+            "timestamp": 1,
+        })
+    )
+
+    kwargs = ch._handle_message.await_args.kwargs
+    assert kwargs["sender_id"] == "5551234"
+
+
+@pytest.mark.asyncio
+async def test_lid_to_phone_cache_resolves_lid_only_messages():
+    """When only LID is present, a cached LID→phone mapping should be used."""
+    ch = WhatsAppChannel({"enabled": True}, MagicMock())
+    ch._handle_message = AsyncMock()
+
+    # First message: both phone and LID → builds cache
+    await ch._handle_bridge_message(
+        json.dumps({
+            "type": "message",
+            "id": "c1",
+            "sender": "LID99@lid.whatsapp.net",
+            "pn": "5559999@s.whatsapp.net",
+            "content": "first",
+            "timestamp": 1,
+        })
+    )
+    # Second message: only LID, no phone
+    await ch._handle_bridge_message(
+        json.dumps({
+            "type": "message",
+            "id": "c2",
+            "sender": "LID99@lid.whatsapp.net",
+            "pn": "",
+            "content": "second",
+            "timestamp": 2,
+        })
+    )
+
+    second_kwargs = ch._handle_message.await_args_list[1].kwargs
+    assert second_kwargs["sender_id"] == "5559999"
+
+
+@pytest.mark.asyncio
+async def test_voice_message_transcription_uses_media_path():
+    """Voice messages are transcribed when media path is available."""
+    ch = WhatsAppChannel({"enabled": True}, MagicMock())
+    ch.transcription_provider = "openai"
+    ch.transcription_api_key = "sk-test"
+    ch._handle_message = AsyncMock()
+    ch.transcribe_audio = AsyncMock(return_value="Hello world")
+
+    await ch._handle_bridge_message(
+        json.dumps({
+            "type": "message",
+            "id": "v1",
+            "sender": "12345@s.whatsapp.net",
+            "pn": "",
+            "content": "[Voice Message]",
+            "timestamp": 1,
+            "media": ["/tmp/voice.ogg"],
+        })
+    )
+
+    ch.transcribe_audio.assert_awaited_once_with("/tmp/voice.ogg")
+    kwargs = ch._handle_message.await_args.kwargs
+    assert kwargs["content"].startswith("Hello world")
+
+
+@pytest.mark.asyncio
+async def test_voice_message_no_media_shows_not_available():
+    """Voice messages without media produce a fallback placeholder."""
+    ch = WhatsAppChannel({"enabled": True}, MagicMock())
+    ch._handle_message = AsyncMock()
+
+    await ch._handle_bridge_message(
+        json.dumps({
+            "type": "message",
+            "id": "v2",
+            "sender": "12345@s.whatsapp.net",
+            "pn": "",
+            "content": "[Voice Message]",
+            "timestamp": 1,
+        })
+    )
+
+    kwargs = ch._handle_message.await_args.kwargs
+    assert kwargs["content"] == "[Voice Message: Audio not available]"
+
+
+def test_load_or_create_bridge_token_persists_generated_secret(tmp_path):
+    token_path = tmp_path / "whatsapp-auth" / "bridge-token"
+
+    first = _load_or_create_bridge_token(token_path)
+    second = _load_or_create_bridge_token(token_path)
+
+    assert first == second
+    assert token_path.read_text(encoding="utf-8") == first
+    assert len(first) >= 32
+    if os.name != "nt":
+        assert token_path.stat().st_mode & 0o777 == 0o600
+
+
+def test_configured_bridge_token_skips_local_token_file(monkeypatch, tmp_path):
+    token_path = tmp_path / "whatsapp-auth" / "bridge-token"
+    monkeypatch.setattr("nanobot.channels.whatsapp._bridge_token_path", lambda: token_path)
+    ch = WhatsAppChannel({"enabled": True, "bridgeToken": "manual-secret"}, MagicMock())
+
+    assert ch._effective_bridge_token() == "manual-secret"
+    assert not token_path.exists()
+
+
+@pytest.mark.asyncio
+async def test_login_exports_effective_bridge_token(monkeypatch, tmp_path):
+    token_path = tmp_path / "whatsapp-auth" / "bridge-token"
+    bridge_dir = tmp_path / "bridge"
+    bridge_dir.mkdir()
+    calls = []
+
+    monkeypatch.setattr("nanobot.channels.whatsapp._bridge_token_path", lambda: token_path)
+    monkeypatch.setattr("nanobot.channels.whatsapp._ensure_bridge_setup", lambda: bridge_dir)
+    monkeypatch.setattr("nanobot.channels.whatsapp.shutil.which", lambda _: "/usr/bin/npm")
+
+    def fake_run(*args, **kwargs):
+        calls.append((args, kwargs))
+        return MagicMock()
+
+    monkeypatch.setattr("nanobot.channels.whatsapp.subprocess.run", fake_run)
+    ch = WhatsAppChannel({"enabled": True}, MagicMock())
+
+    assert await ch.login() is True
+    assert len(calls) == 1
+
+    _, kwargs = calls[0]
+    assert kwargs["cwd"] == bridge_dir
+    assert kwargs["env"]["AUTH_DIR"] == str(token_path.parent)
+    assert kwargs["env"]["BRIDGE_TOKEN"] == token_path.read_text(encoding="utf-8")
+
+
+@pytest.mark.asyncio
+async def test_start_sends_auth_message_with_generated_token(monkeypatch, tmp_path):
+    token_path = tmp_path / "whatsapp-auth" / "bridge-token"
+    sent_messages: list[str] = []
+
+    class FakeWS:
+        def __init__(self) -> None:
+            self.close = AsyncMock()
+
+        async def send(self, message: str) -> None:
+            sent_messages.append(message)
+            ch._running = False
+
+        def __aiter__(self):
+            return self
+
+        async def __anext__(self):
+            raise StopAsyncIteration
+
+    class FakeConnect:
+        def __init__(self, ws):
+            self.ws = ws
+
+        async def __aenter__(self):
+            return self.ws
+
+        async def __aexit__(self, exc_type, exc, tb):
+            return False
+
+    monkeypatch.setattr("nanobot.channels.whatsapp._bridge_token_path", lambda: token_path)
+    monkeypatch.setitem(
+        sys.modules,
+        "websockets",
+        types.SimpleNamespace(connect=lambda url: FakeConnect(FakeWS())),
+    )
+
+    ch = WhatsAppChannel({"enabled": True, "bridgeUrl": "ws://localhost:3001"}, MagicMock())
+    await ch.start()
+
+    assert sent_messages == [
+        json.dumps({"type": "auth", "token": token_path.read_text(encoding="utf-8")})
+    ]

tests/cli/test_cli_input.py

@@ -145,3 +145,29 @@ def test_response_renderable_without_metadata_keeps_markdown_path():
     renderable = commands._response_renderable(help_text, render_markdown=True)
 
     assert renderable.__class__.__name__ == "Markdown"
+
+
+def test_stream_renderer_stop_for_input_stops_spinner():
+    """stop_for_input should stop the active spinner to avoid prompt_toolkit conflicts."""
+    spinner = MagicMock()
+    mock_console = MagicMock()
+    mock_console.status.return_value = spinner
+
+    # Create renderer with mocked console
+    with patch.object(stream_mod, "_make_console", return_value=mock_console):
+        renderer = stream_mod.StreamRenderer(show_spinner=True)
+        
+        # Verify spinner started
+        spinner.start.assert_called_once()
+        
+        # Stop for input
+        renderer.stop_for_input()
+        
+        # Verify spinner stopped
+        spinner.stop.assert_called_once()
+
+
+def test_make_console_uses_force_terminal():
+    """Console should be created with force_terminal=True for proper ANSI handling."""
+    console = stream_mod._make_console()
+    assert console._force_terminal is True

tests/cli/test_commands.py

@@ -1,5 +1,7 @@
+import asyncio
 import json
 import re
+import shutil
 from pathlib import Path
 from unittest.mock import AsyncMock, MagicMock, patch
 
@@ -9,6 +11,7 @@ from typer.testing import CliRunner
 from nanobot.bus.events import OutboundMessage
 from nanobot.cli.commands import _make_provider, app
 from nanobot.config.schema import Config
+from nanobot.cron.types import CronJob, CronPayload
 from nanobot.providers.openai_codex_provider import _strip_model_prefix
 from nanobot.providers.registry import find_by_name
 
@@ -19,11 +22,6 @@ class _StopGatewayError(RuntimeError):
     pass
 
 
-import shutil
-
-import pytest
-
-
 @pytest.fixture
 def mock_paths():
     """Mock config/workspace paths for test isolation."""
@@ -31,7 +29,6 @@ def mock_paths():
          patch("nanobot.config.loader.save_config") as mock_sc, \
          patch("nanobot.config.loader.load_config") as mock_lc, \
          patch("nanobot.cli.commands.get_workspace_path") as mock_ws:
-
         base_dir = Path("./test_onboard_data")
         if base_dir.exists():
             shutil.rmtree(base_dir)
@@ -425,13 +422,13 @@ def mock_agent_runtime(tmp_path):
     config.agents.defaults.workspace = str(tmp_path / "default-workspace")
 
     with patch("nanobot.config.loader.load_config", return_value=config) as mock_load_config, \
+         patch("nanobot.config.loader.resolve_config_env_vars", side_effect=lambda c: c), \
          patch("nanobot.cli.commands.sync_workspace_templates") as mock_sync_templates, \
          patch("nanobot.cli.commands._make_provider", return_value=object()), \
          patch("nanobot.cli.commands._print_agent_response") as mock_print_response, \
          patch("nanobot.bus.queue.MessageBus"), \
          patch("nanobot.cron.service.CronService"), \
          patch("nanobot.agent.loop.AgentLoop") as mock_agent_loop_cls:
-
         agent_loop = MagicMock()
         agent_loop.channels_config = None
         agent_loop.process_direct = AsyncMock(
@@ -656,7 +653,9 @@ def test_agent_custom_config_workspace_does_not_migrate_legacy_cron(
 
     monkeypatch.setattr("nanobot.cron.service.CronService", _FakeCron)
     monkeypatch.setattr("nanobot.agent.loop.AgentLoop", _FakeAgentLoop)
-    monkeypatch.setattr("nanobot.cli.commands._print_agent_response", lambda *_args, **_kwargs: None)
+    monkeypatch.setattr(
+        "nanobot.cli.commands._print_agent_response", lambda *_args, **_kwargs: None
+    )
 
     result = runner.invoke(app, ["agent", "-m", "hello", "-c", str(config_file)])
 
@@ -739,6 +738,7 @@ def _patch_cli_command_runtime(
         set_config_path or (lambda _path: None),
     )
     monkeypatch.setattr("nanobot.config.loader.load_config", lambda _path=None: config)
+    monkeypatch.setattr("nanobot.config.loader.resolve_config_env_vars", lambda c: c)
     monkeypatch.setattr(
         "nanobot.cli.commands.sync_workspace_templates",
         sync_templates or (lambda _path: None),
@@ -868,6 +868,115 @@ def test_gateway_uses_workspace_directory_for_cron_store(monkeypatch, tmp_path:
     assert seen["cron_store"] == config.workspace_path / "cron" / "jobs.json"
 
 
+def test_gateway_cron_evaluator_receives_scheduled_reminder_context(
+    monkeypatch, tmp_path: Path
+) -> None:
+    config_file = tmp_path / "instance" / "config.json"
+    config_file.parent.mkdir(parents=True)
+    config_file.write_text("{}")
+
+    config = Config()
+    config.agents.defaults.workspace = str(tmp_path / "config-workspace")
+    provider = object()
+    bus = MagicMock()
+    bus.publish_outbound = AsyncMock()
+    seen: dict[str, object] = {}
+
+    monkeypatch.setattr("nanobot.config.loader.set_config_path", lambda _path: None)
+    monkeypatch.setattr("nanobot.config.loader.load_config", lambda _path=None: config)
+    monkeypatch.setattr("nanobot.cli.commands.sync_workspace_templates", lambda _path: None)
+    monkeypatch.setattr("nanobot.cli.commands._make_provider", lambda _config: provider)
+    monkeypatch.setattr("nanobot.bus.queue.MessageBus", lambda: bus)
+    monkeypatch.setattr("nanobot.session.manager.SessionManager", lambda _workspace: object())
+
+    class _FakeCron:
+        def __init__(self, _store_path: Path) -> None:
+            self.on_job = None
+            seen["cron"] = self
+
+    class _FakeAgentLoop:
+        def __init__(self, *args, **kwargs) -> None:
+            self.model = "test-model"
+            self.tools = {}
+
+        async def process_direct(self, *_args, **_kwargs):
+            return OutboundMessage(
+                channel="telegram",
+                chat_id="user-1",
+                content="Time to stretch.",
+            )
+
+        async def close_mcp(self) -> None:
+            return None
+
+        async def run(self) -> None:
+            return None
+
+        def stop(self) -> None:
+            return None
+
+    class _StopAfterCronSetup:
+        def __init__(self, *_args, **_kwargs) -> None:
+            raise _StopGatewayError("stop")
+
+    async def _capture_evaluate_response(
+        response: str,
+        task_context: str,
+        provider_arg: object,
+        model: str,
+    ) -> bool:
+        seen["response"] = response
+        seen["task_context"] = task_context
+        seen["provider"] = provider_arg
+        seen["model"] = model
+        return True
+
+    monkeypatch.setattr("nanobot.cron.service.CronService", _FakeCron)
+    monkeypatch.setattr("nanobot.agent.loop.AgentLoop", _FakeAgentLoop)
+    monkeypatch.setattr("nanobot.channels.manager.ChannelManager", _StopAfterCronSetup)
+    monkeypatch.setattr(
+        "nanobot.utils.evaluator.evaluate_response",
+        _capture_evaluate_response,
+    )
+
+    result = runner.invoke(app, ["gateway", "--config", str(config_file)])
+
+    assert isinstance(result.exception, _StopGatewayError)
+    cron = seen["cron"]
+    assert isinstance(cron, _FakeCron)
+    assert cron.on_job is not None
+
+    job = CronJob(
+        id="cron-1",
+        name="stretch",
+        payload=CronPayload(
+            message="Remind me to stretch.",
+            deliver=True,
+            channel="telegram",
+            to="user-1",
+        ),
+    )
+
+    response = asyncio.run(cron.on_job(job))
+
+    assert response == "Time to stretch."
+    assert seen["response"] == "Time to stretch."
+    assert seen["provider"] is provider
+    assert seen["model"] == "test-model"
+    assert seen["task_context"] == (
+        "[Scheduled Task] Timer finished.\n\n"
+        "Task 'stretch' has been triggered.\n"
+        "Scheduled instruction: Remind me to stretch."
+    )
+    bus.publish_outbound.assert_awaited_once_with(
+        OutboundMessage(
+            channel="telegram",
+            chat_id="user-1",
+            content="Time to stretch.",
+        )
+    )
+
+
 def test_gateway_workspace_override_does_not_migrate_legacy_cron(
     monkeypatch, tmp_path: Path
 ) -> None:

tests/cli/test_restart_command.py

@@ -137,7 +137,7 @@ class TestRestartCommand:
         loop.sessions.get_or_create.return_value = session
         loop._start_time = time.time() - 125
         loop._last_usage = {"prompt_tokens": 0, "completion_tokens": 0}
-        loop.memory_consolidator.estimate_session_prompt_tokens = MagicMock(
+        loop.consolidator.estimate_session_prompt_tokens = MagicMock(
             return_value=(20500, "tiktoken")
         )
 
@@ -176,7 +176,7 @@ class TestRestartCommand:
         session.get_history.return_value = [{"role": "user"}]
         loop.sessions.get_or_create.return_value = session
         loop._last_usage = {"prompt_tokens": 1200, "completion_tokens": 34}
-        loop.memory_consolidator.estimate_session_prompt_tokens = MagicMock(
+        loop.consolidator.estimate_session_prompt_tokens = MagicMock(
             return_value=(0, "none")
         )
 

tests/command/test_builtin_dream.py

@@ -0,0 +1,143 @@
+from __future__ import annotations
+
+from types import SimpleNamespace
+
+import pytest
+
+from nanobot.bus.events import InboundMessage
+from nanobot.command.builtin import cmd_dream_log, cmd_dream_restore
+from nanobot.command.router import CommandContext
+from nanobot.utils.gitstore import CommitInfo
+
+
+class _FakeStore:
+    def __init__(self, git, last_dream_cursor: int = 1):
+        self.git = git
+        self._last_dream_cursor = last_dream_cursor
+
+    def get_last_dream_cursor(self) -> int:
+        return self._last_dream_cursor
+
+
+class _FakeGit:
+    def __init__(
+        self,
+        *,
+        initialized: bool = True,
+        commits: list[CommitInfo] | None = None,
+        diff_map: dict[str, tuple[CommitInfo, str] | None] | None = None,
+        revert_result: str | None = None,
+    ):
+        self._initialized = initialized
+        self._commits = commits or []
+        self._diff_map = diff_map or {}
+        self._revert_result = revert_result
+
+    def is_initialized(self) -> bool:
+        return self._initialized
+
+    def log(self, max_entries: int = 20) -> list[CommitInfo]:
+        return self._commits[:max_entries]
+
+    def show_commit_diff(self, sha: str, max_entries: int = 20):
+        return self._diff_map.get(sha)
+
+    def revert(self, sha: str) -> str | None:
+        return self._revert_result
+
+
+def _make_ctx(raw: str, git: _FakeGit, *, args: str = "", last_dream_cursor: int = 1) -> CommandContext:
+    msg = InboundMessage(channel="cli", sender_id="u1", chat_id="direct", content=raw)
+    store = _FakeStore(git, last_dream_cursor=last_dream_cursor)
+    loop = SimpleNamespace(consolidator=SimpleNamespace(store=store))
+    return CommandContext(msg=msg, session=None, key=msg.session_key, raw=raw, args=args, loop=loop)
+
+
+@pytest.mark.asyncio
+async def test_dream_log_latest_is_more_user_friendly() -> None:
+    commit = CommitInfo(sha="abcd1234", message="dream: 2026-04-04, 2 change(s)", timestamp="2026-04-04 12:00")
+    diff = (
+        "diff --git a/SOUL.md b/SOUL.md\n"
+        "--- a/SOUL.md\n"
+        "+++ b/SOUL.md\n"
+        "@@ -1 +1 @@\n"
+        "-old\n"
+        "+new\n"
+    )
+    git = _FakeGit(commits=[commit], diff_map={commit.sha: (commit, diff)})
+
+    out = await cmd_dream_log(_make_ctx("/dream-log", git))
+
+    assert "## Dream Update" in out.content
+    assert "Here is the latest Dream memory change." in out.content
+    assert "- Commit: `abcd1234`" in out.content
+    assert "- Changed files: `SOUL.md`" in out.content
+    assert "Use `/dream-restore abcd1234` to undo this change." in out.content
+    assert "```diff" in out.content
+
+
+@pytest.mark.asyncio
+async def test_dream_log_missing_commit_guides_user() -> None:
+    git = _FakeGit(diff_map={})
+
+    out = await cmd_dream_log(_make_ctx("/dream-log deadbeef", git, args="deadbeef"))
+
+    assert "Couldn't find Dream change `deadbeef`." in out.content
+    assert "Use `/dream-restore` to list recent versions" in out.content
+
+
+@pytest.mark.asyncio
+async def test_dream_log_before_first_run_is_clear() -> None:
+    git = _FakeGit(initialized=False)
+
+    out = await cmd_dream_log(_make_ctx("/dream-log", git, last_dream_cursor=0))
+
+    assert "Dream has not run yet." in out.content
+    assert "Run `/dream`" in out.content
+
+
+@pytest.mark.asyncio
+async def test_dream_restore_lists_versions_with_next_steps() -> None:
+    commits = [
+        CommitInfo(sha="abcd1234", message="dream: latest", timestamp="2026-04-04 12:00"),
+        CommitInfo(sha="bbbb2222", message="dream: older", timestamp="2026-04-04 08:00"),
+    ]
+    git = _FakeGit(commits=commits)
+
+    out = await cmd_dream_restore(_make_ctx("/dream-restore", git))
+
+    assert "## Dream Restore" in out.content
+    assert "Choose a Dream memory version to restore." in out.content
+    assert "`abcd1234` 2026-04-04 12:00 - dream: latest" in out.content
+    assert "Preview a version with `/dream-log <sha>`" in out.content
+    assert "Restore a version with `/dream-restore <sha>`." in out.content
+
+
+@pytest.mark.asyncio
+async def test_dream_restore_success_mentions_files_and_followup() -> None:
+    commit = CommitInfo(sha="abcd1234", message="dream: latest", timestamp="2026-04-04 12:00")
+    diff = (
+        "diff --git a/SOUL.md b/SOUL.md\n"
+        "--- a/SOUL.md\n"
+        "+++ b/SOUL.md\n"
+        "@@ -1 +1 @@\n"
+        "-old\n"
+        "+new\n"
+        "diff --git a/memory/MEMORY.md b/memory/MEMORY.md\n"
+        "--- a/memory/MEMORY.md\n"
+        "+++ b/memory/MEMORY.md\n"
+        "@@ -1 +1 @@\n"
+        "-old\n"
+        "+new\n"
+    )
+    git = _FakeGit(
+        diff_map={commit.sha: (commit, diff)},
+        revert_result="eeee9999",
+    )
+
+    out = await cmd_dream_restore(_make_ctx("/dream-restore abcd1234", git, args="abcd1234"))
+
+    assert "Restored Dream memory to the state before `abcd1234`." in out.content
+    assert "- New safety commit: `eeee9999`" in out.content
+    assert "- Restored files: `SOUL.md`, `memory/MEMORY.md`" in out.content
+    assert "Use `/dream-log eeee9999` to inspect the restore diff." in out.content

tests/config/test_config_migration.py

@@ -1,6 +1,18 @@
 import json
+import socket
+from unittest.mock import patch
 
 from nanobot.config.loader import load_config, save_config
+from nanobot.security.network import validate_url_target
+
+
+def _fake_resolve(host: str, results: list[str]):
+    """Return a getaddrinfo mock that maps the given host to fake IP results."""
+    def _resolver(hostname, port, family=0, type_=0):
+        if hostname == host:
+            return [(socket.AF_INET, socket.SOCK_STREAM, 0, "", (ip, 0)) for ip in results]
+        raise socket.gaierror(f"cannot resolve {hostname}")
+    return _resolver
 
 
 def test_load_config_keeps_max_tokens_and_ignores_legacy_memory_window(tmp_path) -> None:
@@ -126,3 +138,23 @@ def test_onboard_refresh_backfills_missing_channel_fields(tmp_path, monkeypatch)
     assert result.exit_code == 0
     saved = json.loads(config_path.read_text(encoding="utf-8"))
     assert saved["channels"]["qq"]["msgFormat"] == "plain"
+
+
+def test_load_config_resets_ssrf_whitelist_when_next_config_is_empty(tmp_path) -> None:
+    whitelisted = tmp_path / "whitelisted.json"
+    whitelisted.write_text(
+        json.dumps({"tools": {"ssrfWhitelist": ["100.64.0.0/10"]}}),
+        encoding="utf-8",
+    )
+    defaulted = tmp_path / "defaulted.json"
+    defaulted.write_text(json.dumps({}), encoding="utf-8")
+
+    load_config(whitelisted)
+    with patch("nanobot.security.network.socket.getaddrinfo", _fake_resolve("ts.local", ["100.100.1.1"])):
+        ok, err = validate_url_target("http://ts.local/api")
+        assert ok, err
+
+    load_config(defaulted)
+    with patch("nanobot.security.network.socket.getaddrinfo", _fake_resolve("ts.local", ["100.100.1.1"])):
+        ok, _ = validate_url_target("http://ts.local/api")
+        assert not ok

tests/config/test_dream_config.py

@@ -0,0 +1,48 @@
+from nanobot.config.schema import DreamConfig
+
+
+def test_dream_config_defaults_to_interval_hours() -> None:
+    cfg = DreamConfig()
+
+    assert cfg.interval_h == 2
+    assert cfg.cron is None
+
+
+def test_dream_config_builds_every_schedule_from_interval() -> None:
+    cfg = DreamConfig(interval_h=3)
+
+    schedule = cfg.build_schedule("UTC")
+
+    assert schedule.kind == "every"
+    assert schedule.every_ms == 3 * 3_600_000
+    assert schedule.expr is None
+
+
+def test_dream_config_honors_legacy_cron_override() -> None:
+    cfg = DreamConfig.model_validate({"cron": "0 */4 * * *"})
+
+    schedule = cfg.build_schedule("UTC")
+
+    assert schedule.kind == "cron"
+    assert schedule.expr == "0 */4 * * *"
+    assert schedule.tz == "UTC"
+    assert cfg.describe_schedule() == "cron 0 */4 * * * (legacy)"
+
+
+def test_dream_config_dump_uses_interval_h_and_hides_legacy_cron() -> None:
+    cfg = DreamConfig.model_validate({"intervalH": 5, "cron": "0 */4 * * *"})
+
+    dumped = cfg.model_dump(by_alias=True)
+
+    assert dumped["intervalH"] == 5
+    assert "cron" not in dumped
+
+
+def test_dream_config_uses_model_override_name_and_accepts_legacy_model() -> None:
+    cfg = DreamConfig.model_validate({"model": "openrouter/sonnet"})
+
+    dumped = cfg.model_dump(by_alias=True)
+
+    assert cfg.model_override == "openrouter/sonnet"
+    assert dumped["modelOverride"] == "openrouter/sonnet"
+    assert "model" not in dumped

tests/config/test_env_interpolation.py

@@ -0,0 +1,82 @@
+import json
+
+import pytest
+
+from nanobot.config.loader import (
+    _resolve_env_vars,
+    load_config,
+    resolve_config_env_vars,
+    save_config,
+)
+
+
+class TestResolveEnvVars:
+    def test_replaces_string_value(self, monkeypatch):
+        monkeypatch.setenv("MY_SECRET", "hunter2")
+        assert _resolve_env_vars("${MY_SECRET}") == "hunter2"
+
+    def test_partial_replacement(self, monkeypatch):
+        monkeypatch.setenv("HOST", "example.com")
+        assert _resolve_env_vars("https://${HOST}/api") == "https://example.com/api"
+
+    def test_multiple_vars_in_one_string(self, monkeypatch):
+        monkeypatch.setenv("USER", "alice")
+        monkeypatch.setenv("PASS", "secret")
+        assert _resolve_env_vars("${USER}:${PASS}") == "alice:secret"
+
+    def test_nested_dicts(self, monkeypatch):
+        monkeypatch.setenv("TOKEN", "abc123")
+        data = {"channels": {"telegram": {"token": "${TOKEN}"}}}
+        result = _resolve_env_vars(data)
+        assert result["channels"]["telegram"]["token"] == "abc123"
+
+    def test_lists(self, monkeypatch):
+        monkeypatch.setenv("VAL", "x")
+        assert _resolve_env_vars(["${VAL}", "plain"]) == ["x", "plain"]
+
+    def test_ignores_non_strings(self):
+        assert _resolve_env_vars(42) == 42
+        assert _resolve_env_vars(True) is True
+        assert _resolve_env_vars(None) is None
+        assert _resolve_env_vars(3.14) == 3.14
+
+    def test_plain_strings_unchanged(self):
+        assert _resolve_env_vars("no vars here") == "no vars here"
+
+    def test_missing_var_raises(self):
+        with pytest.raises(ValueError, match="DOES_NOT_EXIST"):
+            _resolve_env_vars("${DOES_NOT_EXIST}")
+
+
+class TestResolveConfig:
+    def test_resolves_env_vars_in_config(self, tmp_path, monkeypatch):
+        monkeypatch.setenv("TEST_API_KEY", "resolved-key")
+        config_path = tmp_path / "config.json"
+        config_path.write_text(
+            json.dumps(
+                {"providers": {"groq": {"apiKey": "${TEST_API_KEY}"}}}
+            ),
+            encoding="utf-8",
+        )
+
+        raw = load_config(config_path)
+        assert raw.providers.groq.api_key == "${TEST_API_KEY}"
+
+        resolved = resolve_config_env_vars(raw)
+        assert resolved.providers.groq.api_key == "resolved-key"
+
+    def test_save_preserves_templates(self, tmp_path, monkeypatch):
+        monkeypatch.setenv("MY_TOKEN", "real-token")
+        config_path = tmp_path / "config.json"
+        config_path.write_text(
+            json.dumps(
+                {"channels": {"telegram": {"token": "${MY_TOKEN}"}}}
+            ),
+            encoding="utf-8",
+        )
+
+        raw = load_config(config_path)
+        save_config(raw, config_path)
+
+        saved = json.loads(config_path.read_text(encoding="utf-8"))
+        assert saved["channels"]["telegram"]["token"] == "${MY_TOKEN}"

tests/cron/test_cron_service.py

@@ -4,7 +4,7 @@ import json
 import pytest
 
 from nanobot.cron.service import CronService
-from nanobot.cron.types import CronSchedule
+from nanobot.cron.types import CronJob, CronPayload, CronSchedule
 
 
 def test_add_job_rejects_unknown_timezone(tmp_path) -> None:
@@ -141,3 +141,18 @@ async def test_running_service_honors_external_disable(tmp_path) -> None:
         assert called == []
     finally:
         service.stop()
+
+
+def test_remove_job_refuses_system_jobs(tmp_path) -> None:
+    service = CronService(tmp_path / "cron" / "jobs.json")
+    service.register_system_job(CronJob(
+        id="dream",
+        name="dream",
+        schedule=CronSchedule(kind="cron", expr="0 */2 * * *", tz="UTC"),
+        payload=CronPayload(kind="system_event"),
+    ))
+
+    result = service.remove_job("dream")
+
+    assert result == "protected"
+    assert service.get_job("dream") is not None

tests/cron/test_cron_tool_list.py

@@ -4,7 +4,7 @@ from datetime import datetime, timezone
 
 from nanobot.agent.tools.cron import CronTool
 from nanobot.cron.service import CronService
-from nanobot.cron.types import CronJobState, CronSchedule
+from nanobot.cron.types import CronJob, CronJobState, CronPayload, CronSchedule
 
 
 def _make_tool(tmp_path) -> CronTool:
@@ -262,6 +262,39 @@ def test_list_shows_next_run(tmp_path) -> None:
     assert "(UTC)" in result
 
 
+def test_list_includes_protected_dream_system_job_with_memory_purpose(tmp_path) -> None:
+    tool = _make_tool(tmp_path)
+    tool._cron.register_system_job(CronJob(
+        id="dream",
+        name="dream",
+        schedule=CronSchedule(kind="cron", expr="0 */2 * * *", tz="UTC"),
+        payload=CronPayload(kind="system_event"),
+    ))
+
+    result = tool._list_jobs()
+
+    assert "- dream (id: dream, cron: 0 */2 * * * (UTC))" in result
+    assert "Dream memory consolidation for long-term memory." in result
+    assert "cannot be removed" in result
+
+
+def test_remove_protected_dream_job_returns_clear_feedback(tmp_path) -> None:
+    tool = _make_tool(tmp_path)
+    tool._cron.register_system_job(CronJob(
+        id="dream",
+        name="dream",
+        schedule=CronSchedule(kind="cron", expr="0 */2 * * *", tz="UTC"),
+        payload=CronPayload(kind="system_event"),
+    ))
+
+    result = tool._remove_job("dream")
+
+    assert "Cannot remove job `dream`." in result
+    assert "Dream memory consolidation job for long-term memory" in result
+    assert "cannot be removed" in result
+    assert tool._cron.get_job("dream") is not None
+
+
 def test_add_cron_job_defaults_to_tool_timezone(tmp_path) -> None:
     tool = _make_tool_with_tz(tmp_path, "Asia/Shanghai")
     tool.set_context("telegram", "chat-1")

tests/providers/test_litellm_kwargs.py

@@ -226,7 +226,39 @@ def test_openai_model_passthrough() -> None:
     assert provider.get_default_model() == "gpt-4o"
 
 
-def test_openai_compat_strips_message_level_reasoning_fields() -> None:
+def test_openai_compat_supports_temperature_matches_reasoning_model_rules() -> None:
+    assert OpenAICompatProvider._supports_temperature("gpt-4o") is True
+    assert OpenAICompatProvider._supports_temperature("gpt-5-chat") is False
+    assert OpenAICompatProvider._supports_temperature("o3-mini") is False
+    assert OpenAICompatProvider._supports_temperature("gpt-4o", reasoning_effort="medium") is False
+
+
+def test_openai_compat_build_kwargs_uses_gpt5_safe_parameters() -> None:
+    spec = find_by_name("openai")
+    with patch("nanobot.providers.openai_compat_provider.AsyncOpenAI"):
+        provider = OpenAICompatProvider(
+            api_key="sk-test-key",
+            default_model="gpt-5-chat",
+            spec=spec,
+        )
+
+    kwargs = provider._build_kwargs(
+        messages=[{"role": "user", "content": "hello"}],
+        tools=None,
+        model="gpt-5-chat",
+        max_tokens=4096,
+        temperature=0.7,
+        reasoning_effort=None,
+        tool_choice=None,
+    )
+
+    assert kwargs["model"] == "gpt-5-chat"
+    assert kwargs["max_completion_tokens"] == 4096
+    assert "max_tokens" not in kwargs
+    assert "temperature" not in kwargs
+
+
+def test_openai_compat_preserves_message_level_reasoning_fields() -> None:
     with patch("nanobot.providers.openai_compat_provider.AsyncOpenAI"):
         provider = OpenAICompatProvider()
 
@@ -247,8 +279,8 @@ def test_openai_compat_strips_message_level_reasoning_fields() -> None:
         }
     ])
 
-    assert "reasoning_content" not in sanitized[0]
-    assert "extra_content" not in sanitized[0]
+    assert sanitized[0]["reasoning_content"] == "hidden"
+    assert sanitized[0]["extra_content"] == {"debug": True}
     assert sanitized[0]["tool_calls"][0]["extra_content"] == {"google": {"thought_signature": "sig"}}
 
 
@@ -275,3 +307,54 @@ async def test_openai_compat_stream_watchdog_returns_error_on_stall(monkeypatch)
     assert result.finish_reason == "error"
     assert result.content is not None
     assert "stream stalled" in result.content
+
+
+# ---------------------------------------------------------------------------
+# Provider-specific thinking parameters (extra_body)
+# ---------------------------------------------------------------------------
+
+def _build_kwargs_for(provider_name: str, model: str, reasoning_effort=None):
+    spec = find_by_name(provider_name)
+    with patch("nanobot.providers.openai_compat_provider.AsyncOpenAI"):
+        p = OpenAICompatProvider(api_key="k", default_model=model, spec=spec)
+    return p._build_kwargs(
+        messages=[{"role": "user", "content": "hi"}],
+        tools=None, model=model, max_tokens=1024, temperature=0.7,
+        reasoning_effort=reasoning_effort, tool_choice=None,
+    )
+
+
+def test_dashscope_thinking_enabled_with_reasoning_effort() -> None:
+    kw = _build_kwargs_for("dashscope", "qwen3-plus", reasoning_effort="medium")
+    assert kw["extra_body"] == {"enable_thinking": True}
+
+
+def test_dashscope_thinking_disabled_for_minimal() -> None:
+    kw = _build_kwargs_for("dashscope", "qwen3-plus", reasoning_effort="minimal")
+    assert kw["extra_body"] == {"enable_thinking": False}
+
+
+def test_dashscope_no_extra_body_when_reasoning_effort_none() -> None:
+    kw = _build_kwargs_for("dashscope", "qwen-turbo", reasoning_effort=None)
+    assert "extra_body" not in kw
+
+
+def test_volcengine_thinking_enabled() -> None:
+    kw = _build_kwargs_for("volcengine", "doubao-seed-2-0-pro", reasoning_effort="high")
+    assert kw["extra_body"] == {"thinking": {"type": "enabled"}}
+
+
+def test_byteplus_thinking_disabled_for_minimal() -> None:
+    kw = _build_kwargs_for("byteplus", "doubao-seed-2-0-pro", reasoning_effort="minimal")
+    assert kw["extra_body"] == {"thinking": {"type": "disabled"}}
+
+
+def test_byteplus_no_extra_body_when_reasoning_effort_none() -> None:
+    kw = _build_kwargs_for("byteplus", "doubao-seed-2-0-pro", reasoning_effort=None)
+    assert "extra_body" not in kw
+
+
+def test_openai_no_thinking_extra_body() -> None:
+    """Non-thinking providers should never get extra_body for thinking."""
+    kw = _build_kwargs_for("openai", "gpt-4o", reasoning_effort="medium")
+    assert "extra_body" not in kw

tests/providers/test_prompt_cache_markers.py

@@ -0,0 +1,87 @@
+from __future__ import annotations
+
+from typing import Any
+
+from nanobot.providers.anthropic_provider import AnthropicProvider
+from nanobot.providers.openai_compat_provider import OpenAICompatProvider
+
+
+def _openai_tools(*names: str) -> list[dict[str, Any]]:
+    return [
+        {
+            "type": "function",
+            "function": {
+                "name": name,
+                "description": f"{name} tool",
+                "parameters": {"type": "object", "properties": {}},
+            },
+        }
+        for name in names
+    ]
+
+
+def _anthropic_tools(*names: str) -> list[dict[str, Any]]:
+    return [
+        {
+            "name": name,
+            "description": f"{name} tool",
+            "input_schema": {"type": "object", "properties": {}},
+        }
+        for name in names
+    ]
+
+
+def _marked_openai_tool_names(tools: list[dict[str, Any]] | None) -> list[str]:
+    if not tools:
+        return []
+    marked: list[str] = []
+    for tool in tools:
+        if "cache_control" in tool:
+            marked.append((tool.get("function") or {}).get("name", ""))
+    return marked
+
+
+def _marked_anthropic_tool_names(tools: list[dict[str, Any]] | None) -> list[str]:
+    if not tools:
+        return []
+    return [tool.get("name", "") for tool in tools if "cache_control" in tool]
+
+
+def test_openai_compat_marks_builtin_boundary_and_tail_tool() -> None:
+    messages = [
+        {"role": "system", "content": "system"},
+        {"role": "assistant", "content": "assistant"},
+        {"role": "user", "content": "user"},
+    ]
+    _, marked_tools = OpenAICompatProvider._apply_cache_control(
+        messages,
+        _openai_tools("read_file", "write_file", "mcp_fs_ls", "mcp_git_status"),
+    )
+    assert _marked_openai_tool_names(marked_tools) == ["write_file", "mcp_git_status"]
+
+
+def test_anthropic_marks_builtin_boundary_and_tail_tool() -> None:
+    messages = [
+        {"role": "user", "content": "u1"},
+        {"role": "assistant", "content": "a1"},
+        {"role": "user", "content": "u2"},
+    ]
+    _, _, marked_tools = AnthropicProvider._apply_cache_control(
+        "system",
+        messages,
+        _anthropic_tools("read_file", "write_file", "mcp_fs_ls", "mcp_git_status"),
+    )
+    assert _marked_anthropic_tool_names(marked_tools) == ["write_file", "mcp_git_status"]
+
+
+def test_openai_compat_marks_only_tail_without_mcp() -> None:
+    messages = [
+        {"role": "system", "content": "system"},
+        {"role": "assistant", "content": "assistant"},
+        {"role": "user", "content": "user"},
+    ]
+    _, marked_tools = OpenAICompatProvider._apply_cache_control(
+        messages,
+        _openai_tools("read_file", "write_file"),
+    )
+    assert _marked_openai_tool_names(marked_tools) == ["write_file"]

tests/security/test_security_network.py

@@ -7,7 +7,7 @@ from unittest.mock import patch
 
 import pytest
 
-from nanobot.security.network import contains_internal_url, validate_url_target
+from nanobot.security.network import configure_ssrf_whitelist, contains_internal_url, validate_url_target
 
 
 def _fake_resolve(host: str, results: list[str]):
@@ -99,3 +99,47 @@ def test_allows_normal_curl():
 
 def test_no_urls_returns_false():
     assert not contains_internal_url("echo hello && ls -la")
+
+
+# ---------------------------------------------------------------------------
+# SSRF whitelist — allow specific CIDR ranges (#2669)
+# ---------------------------------------------------------------------------
+
+def test_blocks_cgnat_by_default():
+    """100.64.0.0/10 (CGNAT / Tailscale) is blocked by default."""
+    with patch("nanobot.security.network.socket.getaddrinfo", _fake_resolve("ts.local", ["100.100.1.1"])):
+        ok, _ = validate_url_target("http://ts.local/api")
+        assert not ok
+
+
+def test_whitelist_allows_cgnat():
+    """Whitelisting 100.64.0.0/10 lets Tailscale addresses through."""
+    configure_ssrf_whitelist(["100.64.0.0/10"])
+    try:
+        with patch("nanobot.security.network.socket.getaddrinfo", _fake_resolve("ts.local", ["100.100.1.1"])):
+            ok, err = validate_url_target("http://ts.local/api")
+            assert ok, f"Whitelisted CGNAT should be allowed, got: {err}"
+    finally:
+        configure_ssrf_whitelist([])
+
+
+def test_whitelist_does_not_affect_other_blocked():
+    """Whitelisting CGNAT must not unblock other private ranges."""
+    configure_ssrf_whitelist(["100.64.0.0/10"])
+    try:
+        with patch("nanobot.security.network.socket.getaddrinfo", _fake_resolve("evil.com", ["10.0.0.1"])):
+            ok, _ = validate_url_target("http://evil.com/secret")
+            assert not ok
+    finally:
+        configure_ssrf_whitelist([])
+
+
+def test_whitelist_invalid_cidr_ignored():
+    """Invalid CIDR entries are silently skipped."""
+    configure_ssrf_whitelist(["not-a-cidr", "100.64.0.0/10"])
+    try:
+        with patch("nanobot.security.network.socket.getaddrinfo", _fake_resolve("ts.local", ["100.100.1.1"])):
+            ok, _ = validate_url_target("http://ts.local/api")
+            assert ok
+    finally:
+        configure_ssrf_whitelist([])

tests/tools/test_exec_env.py

@@ -0,0 +1,38 @@
+"""Tests for exec tool environment isolation."""
+
+import pytest
+
+from nanobot.agent.tools.shell import ExecTool
+
+
+@pytest.mark.asyncio
+async def test_exec_does_not_leak_parent_env(monkeypatch):
+    """Env vars from the parent process must not be visible to commands."""
+    monkeypatch.setenv("NANOBOT_SECRET_TOKEN", "super-secret-value")
+    tool = ExecTool()
+    result = await tool.execute(command="printenv NANOBOT_SECRET_TOKEN")
+    assert "super-secret-value" not in result
+
+
+@pytest.mark.asyncio
+async def test_exec_has_working_path():
+    """Basic commands should be available via the login shell's PATH."""
+    tool = ExecTool()
+    result = await tool.execute(command="echo hello")
+    assert "hello" in result
+
+
+@pytest.mark.asyncio
+async def test_exec_path_append():
+    """The pathAppend config should be available in the command's PATH."""
+    tool = ExecTool(path_append="/opt/custom/bin")
+    result = await tool.execute(command="echo $PATH")
+    assert "/opt/custom/bin" in result
+
+
+@pytest.mark.asyncio
+async def test_exec_path_append_preserves_system_path():
+    """pathAppend must not clobber standard system paths."""
+    tool = ExecTool(path_append="/opt/custom/bin")
+    result = await tool.execute(command="ls /")
+    assert "Exit code: 0" in result

tests/tools/test_sandbox.py

@@ -0,0 +1,121 @@
+"""Tests for nanobot.agent.tools.sandbox."""
+
+import shlex
+
+import pytest
+
+from nanobot.agent.tools.sandbox import wrap_command
+
+
+def _parse(cmd: str) -> list[str]:
+    """Split a wrapped command back into tokens for assertion."""
+    return shlex.split(cmd)
+
+
+class TestBwrapBackend:
+    def test_basic_structure(self, tmp_path):
+        ws = str(tmp_path / "project")
+        result = wrap_command("bwrap", "echo hi", ws, ws)
+        tokens = _parse(result)
+
+        assert tokens[0] == "bwrap"
+        assert "--new-session" in tokens
+        assert "--die-with-parent" in tokens
+        assert "--ro-bind" in tokens
+        assert "--proc" in tokens
+        assert "--dev" in tokens
+        assert "--tmpfs" in tokens
+
+        sep = tokens.index("--")
+        assert tokens[sep + 1:] == ["sh", "-c", "echo hi"]
+
+    def test_workspace_bind_mounted_rw(self, tmp_path):
+        ws = str(tmp_path / "project")
+        result = wrap_command("bwrap", "ls", ws, ws)
+        tokens = _parse(result)
+
+        bind_idx = [i for i, t in enumerate(tokens) if t == "--bind"]
+        assert any(tokens[i + 1] == ws and tokens[i + 2] == ws for i in bind_idx)
+
+    def test_parent_dir_masked_with_tmpfs(self, tmp_path):
+        ws = tmp_path / "project"
+        result = wrap_command("bwrap", "ls", str(ws), str(ws))
+        tokens = _parse(result)
+
+        tmpfs_indices = [i for i, t in enumerate(tokens) if t == "--tmpfs"]
+        tmpfs_targets = {tokens[i + 1] for i in tmpfs_indices}
+        assert str(ws.parent) in tmpfs_targets
+
+    def test_cwd_inside_workspace(self, tmp_path):
+        ws = tmp_path / "project"
+        sub = ws / "src" / "lib"
+        result = wrap_command("bwrap", "pwd", str(ws), str(sub))
+        tokens = _parse(result)
+
+        chdir_idx = tokens.index("--chdir")
+        assert tokens[chdir_idx + 1] == str(sub)
+
+    def test_cwd_outside_workspace_falls_back(self, tmp_path):
+        ws = tmp_path / "project"
+        outside = tmp_path / "other"
+        result = wrap_command("bwrap", "pwd", str(ws), str(outside))
+        tokens = _parse(result)
+
+        chdir_idx = tokens.index("--chdir")
+        assert tokens[chdir_idx + 1] == str(ws.resolve())
+
+    def test_command_with_special_characters(self, tmp_path):
+        ws = str(tmp_path / "project")
+        cmd = "echo 'hello world' && cat \"file with spaces.txt\""
+        result = wrap_command("bwrap", cmd, ws, ws)
+        tokens = _parse(result)
+
+        sep = tokens.index("--")
+        assert tokens[sep + 1:] == ["sh", "-c", cmd]
+
+    def test_system_dirs_ro_bound(self, tmp_path):
+        ws = str(tmp_path / "project")
+        result = wrap_command("bwrap", "ls", ws, ws)
+        tokens = _parse(result)
+
+        ro_bind_indices = [i for i, t in enumerate(tokens) if t == "--ro-bind"]
+        ro_targets = {tokens[i + 1] for i in ro_bind_indices}
+        assert "/usr" in ro_targets
+
+    def test_optional_dirs_use_ro_bind_try(self, tmp_path):
+        ws = str(tmp_path / "project")
+        result = wrap_command("bwrap", "ls", ws, ws)
+        tokens = _parse(result)
+
+        try_indices = [i for i, t in enumerate(tokens) if t == "--ro-bind-try"]
+        try_targets = {tokens[i + 1] for i in try_indices}
+        assert "/bin" in try_targets
+        assert "/etc/ssl/certs" in try_targets
+
+    def test_media_dir_ro_bind(self, tmp_path, monkeypatch):
+        """Media directory should be read-only mounted inside the sandbox."""
+        fake_media = tmp_path / "media"
+        fake_media.mkdir()
+        monkeypatch.setattr(
+            "nanobot.agent.tools.sandbox.get_media_dir",
+            lambda: fake_media,
+        )
+        ws = str(tmp_path / "project")
+        result = wrap_command("bwrap", "ls", ws, ws)
+        tokens = _parse(result)
+
+        try_indices = [i for i, t in enumerate(tokens) if t == "--ro-bind-try"]
+        try_pairs = {(tokens[i + 1], tokens[i + 2]) for i in try_indices}
+        assert (str(fake_media), str(fake_media)) in try_pairs
+
+
+class TestUnknownBackend:
+    def test_raises_value_error(self, tmp_path):
+        ws = str(tmp_path / "project")
+        with pytest.raises(ValueError, match="Unknown sandbox backend"):
+            wrap_command("nonexistent", "ls", ws, ws)
+
+    def test_empty_string_raises(self, tmp_path):
+        ws = str(tmp_path / "project")
+        with pytest.raises(ValueError):
+            wrap_command("", "ls", ws, ws)

tests/tools/test_search_tools.py

@@ -0,0 +1,325 @@
+"""Tests for grep/glob search tools."""
+
+from __future__ import annotations
+
+import os
+from pathlib import Path
+from types import SimpleNamespace
+from unittest.mock import AsyncMock, MagicMock
+
+import pytest
+
+from nanobot.agent.loop import AgentLoop
+from nanobot.agent.subagent import SubagentManager
+from nanobot.agent.tools.search import GlobTool, GrepTool
+from nanobot.bus.queue import MessageBus
+
+
+@pytest.mark.asyncio
+async def test_glob_matches_recursively_and_skips_noise_dirs(tmp_path: Path) -> None:
+    (tmp_path / "src").mkdir()
+    (tmp_path / "nested").mkdir()
+    (tmp_path / "node_modules").mkdir()
+    (tmp_path / "src" / "app.py").write_text("print('ok')\n", encoding="utf-8")
+    (tmp_path / "nested" / "util.py").write_text("print('ok')\n", encoding="utf-8")
+    (tmp_path / "node_modules" / "skip.py").write_text("print('skip')\n", encoding="utf-8")
+
+    tool = GlobTool(workspace=tmp_path, allowed_dir=tmp_path)
+    result = await tool.execute(pattern="*.py", path=".")
+
+    assert "src/app.py" in result
+    assert "nested/util.py" in result
+    assert "node_modules/skip.py" not in result
+
+
+@pytest.mark.asyncio
+async def test_glob_can_return_directories_only(tmp_path: Path) -> None:
+    (tmp_path / "src").mkdir()
+    (tmp_path / "src" / "api").mkdir(parents=True)
+    (tmp_path / "src" / "api" / "handlers.py").write_text("ok\n", encoding="utf-8")
+
+    tool = GlobTool(workspace=tmp_path, allowed_dir=tmp_path)
+    result = await tool.execute(
+        pattern="api",
+        path="src",
+        entry_type="dirs",
+    )
+
+    assert result.splitlines() == ["src/api/"]
+
+
+@pytest.mark.asyncio
+async def test_grep_respects_glob_filter_and_context(tmp_path: Path) -> None:
+    (tmp_path / "src").mkdir()
+    (tmp_path / "src" / "main.py").write_text(
+        "alpha\nbeta\nmatch_here\ngamma\n",
+        encoding="utf-8",
+    )
+    (tmp_path / "README.md").write_text("match_here\n", encoding="utf-8")
+
+    tool = GrepTool(workspace=tmp_path, allowed_dir=tmp_path)
+    result = await tool.execute(
+        pattern="match_here",
+        path=".",
+        glob="*.py",
+        output_mode="content",
+        context_before=1,
+        context_after=1,
+    )
+
+    assert "src/main.py:3" in result
+    assert "  2| beta" in result
+    assert "> 3| match_here" in result
+    assert "  4| gamma" in result
+    assert "README.md" not in result
+
+
+@pytest.mark.asyncio
+async def test_grep_defaults_to_files_with_matches(tmp_path: Path) -> None:
+    (tmp_path / "src").mkdir()
+    (tmp_path / "src" / "main.py").write_text("match_here\n", encoding="utf-8")
+
+    tool = GrepTool(workspace=tmp_path, allowed_dir=tmp_path)
+    result = await tool.execute(
+        pattern="match_here",
+        path="src",
+    )
+
+    assert result.splitlines() == ["src/main.py"]
+    assert "1|" not in result
+
+
+@pytest.mark.asyncio
+async def test_grep_supports_case_insensitive_search(tmp_path: Path) -> None:
+    (tmp_path / "memory").mkdir()
+    (tmp_path / "memory" / "HISTORY.md").write_text(
+        "[2026-04-02 10:00] OAuth token rotated\n",
+        encoding="utf-8",
+    )
+
+    tool = GrepTool(workspace=tmp_path, allowed_dir=tmp_path)
+    result = await tool.execute(
+        pattern="oauth",
+        path="memory/HISTORY.md",
+        case_insensitive=True,
+        output_mode="content",
+    )
+
+    assert "memory/HISTORY.md:1" in result
+    assert "OAuth token rotated" in result
+
+
+@pytest.mark.asyncio
+async def test_grep_type_filter_limits_files(tmp_path: Path) -> None:
+    (tmp_path / "src").mkdir()
+    (tmp_path / "src" / "a.py").write_text("needle\n", encoding="utf-8")
+    (tmp_path / "src" / "b.md").write_text("needle\n", encoding="utf-8")
+
+    tool = GrepTool(workspace=tmp_path, allowed_dir=tmp_path)
+    result = await tool.execute(
+        pattern="needle",
+        path="src",
+        type="py",
+    )
+
+    assert result.splitlines() == ["src/a.py"]
+
+
+@pytest.mark.asyncio
+async def test_grep_fixed_strings_treats_regex_chars_literally(tmp_path: Path) -> None:
+    (tmp_path / "memory").mkdir()
+    (tmp_path / "memory" / "HISTORY.md").write_text(
+        "[2026-04-02 10:00] OAuth token rotated\n",
+        encoding="utf-8",
+    )
+
+    tool = GrepTool(workspace=tmp_path, allowed_dir=tmp_path)
+    result = await tool.execute(
+        pattern="[2026-04-02 10:00]",
+        path="memory/HISTORY.md",
+        fixed_strings=True,
+        output_mode="content",
+    )
+
+    assert "memory/HISTORY.md:1" in result
+    assert "[2026-04-02 10:00] OAuth token rotated" in result
+
+
+@pytest.mark.asyncio
+async def test_grep_files_with_matches_mode_returns_unique_paths(tmp_path: Path) -> None:
+    (tmp_path / "src").mkdir()
+    a = tmp_path / "src" / "a.py"
+    b = tmp_path / "src" / "b.py"
+    a.write_text("needle\nneedle\n", encoding="utf-8")
+    b.write_text("needle\n", encoding="utf-8")
+    os.utime(a, (1, 1))
+    os.utime(b, (2, 2))
+
+    tool = GrepTool(workspace=tmp_path, allowed_dir=tmp_path)
+    result = await tool.execute(
+        pattern="needle",
+        path="src",
+        output_mode="files_with_matches",
+    )
+
+    assert result.splitlines() == ["src/b.py", "src/a.py"]
+
+
+@pytest.mark.asyncio
+async def test_grep_files_with_matches_supports_head_limit_and_offset(tmp_path: Path) -> None:
+    (tmp_path / "src").mkdir()
+    for name in ("a.py", "b.py", "c.py"):
+        (tmp_path / "src" / name).write_text("needle\n", encoding="utf-8")
+
+    tool = GrepTool(workspace=tmp_path, allowed_dir=tmp_path)
+    result = await tool.execute(
+        pattern="needle",
+        path="src",
+        head_limit=1,
+        offset=1,
+    )
+
+    lines = result.splitlines()
+    assert lines[0] == "src/b.py"
+    assert "pagination: limit=1, offset=1" in result
+
+
+@pytest.mark.asyncio
+async def test_grep_count_mode_reports_counts_per_file(tmp_path: Path) -> None:
+    (tmp_path / "logs").mkdir()
+    (tmp_path / "logs" / "one.log").write_text("warn\nok\nwarn\n", encoding="utf-8")
+    (tmp_path / "logs" / "two.log").write_text("warn\n", encoding="utf-8")
+
+    tool = GrepTool(workspace=tmp_path, allowed_dir=tmp_path)
+    result = await tool.execute(
+        pattern="warn",
+        path="logs",
+        output_mode="count",
+    )
+
+    assert "logs/one.log: 2" in result
+    assert "logs/two.log: 1" in result
+    assert "total matches: 3 in 2 files" in result
+
+
+@pytest.mark.asyncio
+async def test_grep_files_with_matches_mode_respects_max_results(tmp_path: Path) -> None:
+    (tmp_path / "src").mkdir()
+    files = []
+    for idx, name in enumerate(("a.py", "b.py", "c.py"), start=1):
+        file_path = tmp_path / "src" / name
+        file_path.write_text("needle\n", encoding="utf-8")
+        os.utime(file_path, (idx, idx))
+        files.append(file_path)
+
+    tool = GrepTool(workspace=tmp_path, allowed_dir=tmp_path)
+    result = await tool.execute(
+        pattern="needle",
+        path="src",
+        output_mode="files_with_matches",
+        max_results=2,
+    )
+
+    assert result.splitlines()[:2] == ["src/c.py", "src/b.py"]
+    assert "pagination: limit=2, offset=0" in result
+
+
+@pytest.mark.asyncio
+async def test_glob_supports_head_limit_offset_and_recent_first(tmp_path: Path) -> None:
+    (tmp_path / "src").mkdir()
+    a = tmp_path / "src" / "a.py"
+    b = tmp_path / "src" / "b.py"
+    c = tmp_path / "src" / "c.py"
+    a.write_text("a\n", encoding="utf-8")
+    b.write_text("b\n", encoding="utf-8")
+    c.write_text("c\n", encoding="utf-8")
+
+    os.utime(a, (1, 1))
+    os.utime(b, (2, 2))
+    os.utime(c, (3, 3))
+
+    tool = GlobTool(workspace=tmp_path, allowed_dir=tmp_path)
+    result = await tool.execute(
+        pattern="*.py",
+        path="src",
+        head_limit=1,
+        offset=1,
+    )
+
+    lines = result.splitlines()
+    assert lines[0] == "src/b.py"
+    assert "pagination: limit=1, offset=1" in result
+
+
+@pytest.mark.asyncio
+async def test_grep_reports_skipped_binary_and_large_files(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    (tmp_path / "binary.bin").write_bytes(b"\x00\x01\x02")
+    (tmp_path / "large.txt").write_text("x" * 20, encoding="utf-8")
+
+    monkeypatch.setattr(GrepTool, "_MAX_FILE_BYTES", 10)
+    tool = GrepTool(workspace=tmp_path, allowed_dir=tmp_path)
+    result = await tool.execute(pattern="needle", path=".")
+
+    assert "No matches found" in result
+    assert "skipped 1 binary/unreadable files" in result
+    assert "skipped 1 large files" in result
+
+
+@pytest.mark.asyncio
+async def test_search_tools_reject_paths_outside_workspace(tmp_path: Path) -> None:
+    outside = tmp_path.parent / "outside-search.txt"
+    outside.write_text("secret\n", encoding="utf-8")
+
+    grep_tool = GrepTool(workspace=tmp_path, allowed_dir=tmp_path)
+    glob_tool = GlobTool(workspace=tmp_path, allowed_dir=tmp_path)
+
+    grep_result = await grep_tool.execute(pattern="secret", path=str(outside))
+    glob_result = await glob_tool.execute(pattern="*.txt", path=str(outside.parent))
+
+    assert grep_result.startswith("Error:")
+    assert glob_result.startswith("Error:")
+
+
+def test_agent_loop_registers_grep_and_glob(tmp_path: Path) -> None:
+    bus = MessageBus()
+    provider = MagicMock()
+    provider.get_default_model.return_value = "test-model"
+
+    loop = AgentLoop(bus=bus, provider=provider, workspace=tmp_path, model="test-model")
+
+    assert "grep" in loop.tools.tool_names
+    assert "glob" in loop.tools.tool_names
+
+
+@pytest.mark.asyncio
+async def test_subagent_registers_grep_and_glob(tmp_path: Path) -> None:
+    bus = MessageBus()
+    provider = MagicMock()
+    provider.get_default_model.return_value = "test-model"
+    mgr = SubagentManager(
+        provider=provider,
+        workspace=tmp_path,
+        bus=bus,
+        max_tool_result_chars=4096,
+    )
+    captured: dict[str, list[str]] = {}
+
+    async def fake_run(spec):
+        captured["tool_names"] = spec.tools.tool_names
+        return SimpleNamespace(
+            stop_reason="ok",
+            final_content="done",
+            tool_events=[],
+            error=None,
+        )
+
+    mgr.runner.run = fake_run
+    mgr._announce_result = AsyncMock()
+
+    await mgr._run_subagent("sub-1", "search task", "label", {"channel": "cli", "chat_id": "direct"})
+
+    assert "grep" in captured["tool_names"]
+    assert "glob" in captured["tool_names"]

tests/tools/test_tool_registry.py

@@ -0,0 +1,49 @@
+from __future__ import annotations
+
+from typing import Any
+
+from nanobot.agent.tools.base import Tool
+from nanobot.agent.tools.registry import ToolRegistry
+
+
+class _FakeTool(Tool):
+    def __init__(self, name: str):
+        self._name = name
+
+    @property
+    def name(self) -> str:
+        return self._name
+
+    @property
+    def description(self) -> str:
+        return f"{self._name} tool"
+
+    @property
+    def parameters(self) -> dict[str, Any]:
+        return {"type": "object", "properties": {}}
+
+    async def execute(self, **kwargs: Any) -> Any:
+        return kwargs
+
+
+def _tool_names(definitions: list[dict[str, Any]]) -> list[str]:
+    names: list[str] = []
+    for definition in definitions:
+        fn = definition.get("function", {})
+        names.append(fn.get("name", ""))
+    return names
+
+
+def test_get_definitions_orders_builtins_then_mcp_tools() -> None:
+    registry = ToolRegistry()
+    registry.register(_FakeTool("mcp_git_status"))
+    registry.register(_FakeTool("write_file"))
+    registry.register(_FakeTool("mcp_fs_list"))
+    registry.register(_FakeTool("read_file"))
+
+    assert _tool_names(registry.get_definitions()) == [
+        "read_file",
+        "write_file",
+        "mcp_fs_list",
+        "mcp_git_status",
+    ]

tests/tools/test_tool_validation.py

@@ -1,5 +1,17 @@
+import shlex
+import subprocess
+import sys
 from typing import Any
 
+from nanobot.agent.tools import (
+    ArraySchema,
+    IntegerSchema,
+    ObjectSchema,
+    Schema,
+    StringSchema,
+    tool_parameters,
+    tool_parameters_schema,
+)
 from nanobot.agent.tools.base import Tool
 from nanobot.agent.tools.registry import ToolRegistry
 from nanobot.agent.tools.shell import ExecTool
@@ -41,6 +53,103 @@ class SampleTool(Tool):
         return "ok"
 
 
+@tool_parameters(
+    tool_parameters_schema(
+        query=StringSchema(min_length=2),
+        count=IntegerSchema(2, minimum=1, maximum=10),
+        required=["query", "count"],
+    )
+)
+class DecoratedSampleTool(Tool):
+    @property
+    def name(self) -> str:
+        return "decorated_sample"
+
+    @property
+    def description(self) -> str:
+        return "decorated sample tool"
+
+    async def execute(self, **kwargs: Any) -> str:
+        return f"ok:{kwargs['count']}"
+
+
+def test_schema_validate_value_matches_tool_validate_params() -> None:
+    """ObjectSchema.validate_value 与 validate_json_schema_value、Tool.validate_params 一致。"""
+    root = tool_parameters_schema(
+        query=StringSchema(min_length=2),
+        count=IntegerSchema(2, minimum=1, maximum=10),
+        required=["query", "count"],
+    )
+    obj = ObjectSchema(
+        query=StringSchema(min_length=2),
+        count=IntegerSchema(2, minimum=1, maximum=10),
+        required=["query", "count"],
+    )
+    params = {"query": "h", "count": 2}
+
+    class _Mini(Tool):
+        @property
+        def name(self) -> str:
+            return "m"
+
+        @property
+        def description(self) -> str:
+            return ""
+
+        @property
+        def parameters(self) -> dict[str, Any]:
+            return root
+
+        async def execute(self, **kwargs: Any) -> str:
+            return ""
+
+    expected = _Mini().validate_params(params)
+    assert Schema.validate_json_schema_value(params, root, "") == expected
+    assert obj.validate_value(params, "") == expected
+    assert IntegerSchema(0, minimum=1).validate_value(0, "n") == ["n must be >= 1"]
+
+
+def test_schema_classes_equivalent_to_sample_tool_parameters() -> None:
+    """Schema 类生成的 JSON Schema 应与手写 dict 一致，便于校验行为一致。"""
+    built = tool_parameters_schema(
+        query=StringSchema(min_length=2),
+        count=IntegerSchema(2, minimum=1, maximum=10),
+        mode=StringSchema("", enum=["fast", "full"]),
+        meta=ObjectSchema(
+            tag=StringSchema(""),
+            flags=ArraySchema(StringSchema("")),
+            required=["tag"],
+        ),
+        required=["query", "count"],
+    )
+    assert built == SampleTool().parameters
+
+
+def test_tool_parameters_returns_fresh_copy_per_access() -> None:
+    tool = DecoratedSampleTool()
+
+    first = tool.parameters
+    second = tool.parameters
+
+    assert first == second
+    assert first is not second
+    assert first["properties"] is not second["properties"]
+
+    first["properties"]["query"]["minLength"] = 99
+    assert tool.parameters["properties"]["query"]["minLength"] == 2
+
+
+async def test_registry_executes_decorated_tool_end_to_end() -> None:
+    reg = ToolRegistry()
+    reg.register(DecoratedSampleTool())
+
+    ok = await reg.execute("decorated_sample", {"query": "hello", "count": "3"})
+    assert ok == "ok:3"
+
+    err = await reg.execute("decorated_sample", {"query": "h", "count": 3})
+    assert "Invalid parameters" in err
+
+
 def test_validate_params_missing_required() -> None:
     tool = SampleTool()
     errors = tool.validate_params({"query": "hi"})
@@ -440,10 +549,15 @@ async def test_exec_head_tail_truncation() -> None:
     """Long output should preserve both head and tail."""
     tool = ExecTool()
     # Generate output that exceeds _MAX_OUTPUT (10_000 chars)
-    # Use python to generate output to avoid command line length limits
-    result = await tool.execute(
-        command="python -c \"print('A' * 6000 + '\\n' + 'B' * 6000)\""
-    )
+    # Use current interpreter (PATH may not have `python`). ExecTool uses
+    # create_subprocess_shell: POSIX needs shlex.quote; Windows uses cmd.exe
+    # rules, so list2cmdline is appropriate there.
+    script = "print('A' * 6000 + '\\n' + 'B' * 6000)"
+    if sys.platform == "win32":
+        command = subprocess.list2cmdline([sys.executable, "-c", script])
+    else:
+        command = f"{shlex.quote(sys.executable)} -c {shlex.quote(script)}"
+    result = await tool.execute(command=command)
     assert "chars truncated" in result
     # Head portion should start with As
     assert result.startswith("A")

tests/tools/test_web_search_tool.py

@@ -1,5 +1,7 @@
 """Tests for multi-provider web search."""
 
+import asyncio
+
 import httpx
 import pytest
 
@@ -160,3 +162,70 @@ async def test_searxng_invalid_url():
     tool = _tool(provider="searxng", base_url="not-a-url")
     result = await tool.execute(query="test")
     assert "Error" in result
+
+
+@pytest.mark.asyncio
+async def test_jina_422_falls_back_to_duckduckgo(monkeypatch):
+    class MockDDGS:
+        def __init__(self, **kw):
+            pass
+
+        def text(self, query, max_results=5):
+            return [{"title": "Fallback", "href": "https://ddg.example", "body": "DuckDuckGo fallback"}]
+
+    async def mock_get(self, url, **kw):
+        assert "s.jina.ai" in str(url)
+        raise httpx.HTTPStatusError(
+            "422 Unprocessable Entity",
+            request=httpx.Request("GET", str(url)),
+            response=httpx.Response(422, request=httpx.Request("GET", str(url))),
+        )
+
+    monkeypatch.setattr(httpx.AsyncClient, "get", mock_get)
+    monkeypatch.setattr("ddgs.DDGS", MockDDGS)
+
+    tool = _tool(provider="jina", api_key="jina-key")
+    result = await tool.execute(query="test")
+    assert "DuckDuckGo fallback" in result
+
+
+@pytest.mark.asyncio
+async def test_jina_search_uses_path_encoded_query(monkeypatch):
+    calls = {}
+
+    async def mock_get(self, url, **kw):
+        calls["url"] = str(url)
+        calls["params"] = kw.get("params")
+        return _response(json={
+            "data": [{"title": "Jina Result", "url": "https://jina.ai", "content": "AI search"}]
+        })
+
+    monkeypatch.setattr(httpx.AsyncClient, "get", mock_get)
+    tool = _tool(provider="jina", api_key="jina-key")
+    await tool.execute(query="hello world")
+    assert calls["url"].rstrip("/") == "https://s.jina.ai/hello%20world"
+    assert calls["params"] in (None, {})
+
+
+@pytest.mark.asyncio
+async def test_duckduckgo_timeout_returns_error(monkeypatch):
+    """asyncio.wait_for guard should fire when DDG search hangs."""
+    import threading
+    gate = threading.Event()
+
+    class HangingDDGS:
+        def __init__(self, **kw):
+            pass
+
+        def text(self, query, max_results=5):
+            gate.wait(timeout=10)
+            return []
+
+    monkeypatch.setattr("ddgs.DDGS", HangingDDGS)
+    tool = _tool(provider="duckduckgo")
+    tool.config.timeout = 0.2
+    result = await tool.execute(query="test")
+    gate.set()
+    assert "Error" in result
+
+

tests/utils/test_searchusage.py

@@ -0,0 +1,303 @@
+"""Tests for web search provider usage fetching and /status integration."""
+
+from __future__ import annotations
+
+import pytest
+from unittest.mock import AsyncMock, MagicMock, patch
+
+from nanobot.utils.searchusage import (
+    SearchUsageInfo,
+    _parse_tavily_usage,
+    fetch_search_usage,
+)
+from nanobot.utils.helpers import build_status_content
+
+
+# ---------------------------------------------------------------------------
+# SearchUsageInfo.format() tests
+# ---------------------------------------------------------------------------
+
+class TestSearchUsageInfoFormat:
+    def test_unsupported_provider_shows_no_tracking(self):
+        info = SearchUsageInfo(provider="duckduckgo", supported=False)
+        text = info.format()
+        assert "duckduckgo" in text
+        assert "not available" in text
+
+    def test_supported_with_error(self):
+        info = SearchUsageInfo(provider="tavily", supported=True, error="HTTP 401")
+        text = info.format()
+        assert "tavily" in text
+        assert "HTTP 401" in text
+        assert "unavailable" in text
+
+    def test_full_tavily_usage(self):
+        info = SearchUsageInfo(
+            provider="tavily",
+            supported=True,
+            used=142,
+            limit=1000,
+            remaining=858,
+            reset_date="2026-05-01",
+            search_used=120,
+            extract_used=15,
+            crawl_used=7,
+        )
+        text = info.format()
+        assert "tavily" in text
+        assert "142 / 1000" in text
+        assert "858" in text
+        assert "2026-05-01" in text
+        assert "Search: 120" in text
+        assert "Extract: 15" in text
+        assert "Crawl: 7" in text
+
+    def test_usage_without_limit(self):
+        info = SearchUsageInfo(provider="tavily", supported=True, used=50)
+        text = info.format()
+        assert "50 requests" in text
+        assert "/" not in text.split("Usage:")[1].split("\n")[0]
+
+    def test_no_breakdown_when_none(self):
+        info = SearchUsageInfo(
+            provider="tavily", supported=True, used=10, limit=100, remaining=90
+        )
+        text = info.format()
+        assert "Breakdown" not in text
+
+    def test_brave_unsupported(self):
+        info = SearchUsageInfo(provider="brave", supported=False)
+        text = info.format()
+        assert "brave" in text
+        assert "not available" in text
+
+
+# ---------------------------------------------------------------------------
+# _parse_tavily_usage tests
+# ---------------------------------------------------------------------------
+
+class TestParseTavilyUsage:
+    def test_full_response(self):
+        data = {
+            "used": 142,
+            "limit": 1000,
+            "remaining": 858,
+            "reset_date": "2026-05-01",
+            "breakdown": {"search": 120, "extract": 15, "crawl": 7},
+        }
+        info = _parse_tavily_usage(data)
+        assert info.provider == "tavily"
+        assert info.supported is True
+        assert info.used == 142
+        assert info.limit == 1000
+        assert info.remaining == 858
+        assert info.reset_date == "2026-05-01"
+        assert info.search_used == 120
+        assert info.extract_used == 15
+        assert info.crawl_used == 7
+
+    def test_remaining_computed_when_missing(self):
+        data = {"used": 300, "limit": 1000}
+        info = _parse_tavily_usage(data)
+        assert info.remaining == 700
+
+    def test_remaining_not_negative(self):
+        data = {"used": 1100, "limit": 1000}
+        info = _parse_tavily_usage(data)
+        assert info.remaining == 0
+
+    def test_camel_case_reset_date(self):
+        data = {"used": 10, "limit": 100, "resetDate": "2026-06-01"}
+        info = _parse_tavily_usage(data)
+        assert info.reset_date == "2026-06-01"
+
+    def test_empty_response(self):
+        info = _parse_tavily_usage({})
+        assert info.provider == "tavily"
+        assert info.supported is True
+        assert info.used is None
+        assert info.limit is None
+
+    def test_no_breakdown_key(self):
+        data = {"used": 5, "limit": 50}
+        info = _parse_tavily_usage(data)
+        assert info.search_used is None
+        assert info.extract_used is None
+        assert info.crawl_used is None
+
+
+# ---------------------------------------------------------------------------
+# fetch_search_usage routing tests
+# ---------------------------------------------------------------------------
+
+class TestFetchSearchUsageRouting:
+    @pytest.mark.asyncio
+    async def test_duckduckgo_returns_unsupported(self):
+        info = await fetch_search_usage("duckduckgo")
+        assert info.provider == "duckduckgo"
+        assert info.supported is False
+
+    @pytest.mark.asyncio
+    async def test_searxng_returns_unsupported(self):
+        info = await fetch_search_usage("searxng")
+        assert info.supported is False
+
+    @pytest.mark.asyncio
+    async def test_jina_returns_unsupported(self):
+        info = await fetch_search_usage("jina")
+        assert info.supported is False
+
+    @pytest.mark.asyncio
+    async def test_brave_returns_unsupported(self):
+        info = await fetch_search_usage("brave")
+        assert info.provider == "brave"
+        assert info.supported is False
+
+    @pytest.mark.asyncio
+    async def test_unknown_provider_returns_unsupported(self):
+        info = await fetch_search_usage("some_unknown_provider")
+        assert info.supported is False
+
+    @pytest.mark.asyncio
+    async def test_tavily_no_api_key_returns_error(self):
+        with patch.dict("os.environ", {}, clear=True):
+            # Ensure TAVILY_API_KEY is not set
+            import os
+            os.environ.pop("TAVILY_API_KEY", None)
+            info = await fetch_search_usage("tavily", api_key=None)
+        assert info.provider == "tavily"
+        assert info.supported is True
+        assert info.error is not None
+        assert "not configured" in info.error
+
+    @pytest.mark.asyncio
+    async def test_tavily_success(self):
+        mock_response = MagicMock()
+        mock_response.status_code = 200
+        mock_response.json.return_value = {
+            "used": 142,
+            "limit": 1000,
+            "remaining": 858,
+            "reset_date": "2026-05-01",
+            "breakdown": {"search": 120, "extract": 15, "crawl": 7},
+        }
+        mock_response.raise_for_status = MagicMock()
+
+        mock_client = AsyncMock()
+        mock_client.__aenter__ = AsyncMock(return_value=mock_client)
+        mock_client.__aexit__ = AsyncMock(return_value=False)
+        mock_client.get = AsyncMock(return_value=mock_response)
+
+        with patch("httpx.AsyncClient", return_value=mock_client):
+            info = await fetch_search_usage("tavily", api_key="test-key")
+
+        assert info.provider == "tavily"
+        assert info.supported is True
+        assert info.error is None
+        assert info.used == 142
+        assert info.limit == 1000
+        assert info.remaining == 858
+        assert info.reset_date == "2026-05-01"
+        assert info.search_used == 120
+
+    @pytest.mark.asyncio
+    async def test_tavily_http_error(self):
+        import httpx
+
+        mock_response = MagicMock()
+        mock_response.status_code = 401
+        mock_response.raise_for_status.side_effect = httpx.HTTPStatusError(
+            "401", request=MagicMock(), response=mock_response
+        )
+
+        mock_client = AsyncMock()
+        mock_client.__aenter__ = AsyncMock(return_value=mock_client)
+        mock_client.__aexit__ = AsyncMock(return_value=False)
+        mock_client.get = AsyncMock(return_value=mock_response)
+
+        with patch("httpx.AsyncClient", return_value=mock_client):
+            info = await fetch_search_usage("tavily", api_key="bad-key")
+
+        assert info.supported is True
+        assert info.error == "HTTP 401"
+
+    @pytest.mark.asyncio
+    async def test_tavily_network_error(self):
+        import httpx
+
+        mock_client = AsyncMock()
+        mock_client.__aenter__ = AsyncMock(return_value=mock_client)
+        mock_client.__aexit__ = AsyncMock(return_value=False)
+        mock_client.get = AsyncMock(side_effect=httpx.ConnectError("timeout"))
+
+        with patch("httpx.AsyncClient", return_value=mock_client):
+            info = await fetch_search_usage("tavily", api_key="test-key")
+
+        assert info.supported is True
+        assert info.error is not None
+
+    @pytest.mark.asyncio
+    async def test_provider_name_case_insensitive(self):
+        info = await fetch_search_usage("Tavily", api_key=None)
+        assert info.provider == "tavily"
+        assert info.supported is True
+
+
+# ---------------------------------------------------------------------------
+# build_status_content integration tests
+# ---------------------------------------------------------------------------
+
+class TestBuildStatusContentWithSearchUsage:
+    _BASE_KWARGS = dict(
+        version="0.1.0",
+        model="claude-opus-4-5",
+        start_time=1_000_000.0,
+        last_usage={"prompt_tokens": 1000, "completion_tokens": 200},
+        context_window_tokens=65536,
+        session_msg_count=5,
+        context_tokens_estimate=3000,
+    )
+
+    def test_no_search_usage_unchanged(self):
+        """Omitting search_usage_text keeps existing behaviour."""
+        content = build_status_content(**self._BASE_KWARGS)
+        assert "🔍" not in content
+        assert "Web Search" not in content
+
+    def test_search_usage_none_unchanged(self):
+        content = build_status_content(**self._BASE_KWARGS, search_usage_text=None)
+        assert "🔍" not in content
+
+    def test_search_usage_appended(self):
+        usage_text = "🔍 Web Search: tavily\n   Usage: 142 / 1000 requests"
+        content = build_status_content(**self._BASE_KWARGS, search_usage_text=usage_text)
+        assert "🔍 Web Search: tavily" in content
+        assert "142 / 1000" in content
+
+    def test_existing_fields_still_present(self):
+        usage_text = "🔍 Web Search: duckduckgo\n   Usage tracking: not available"
+        content = build_status_content(**self._BASE_KWARGS, search_usage_text=usage_text)
+        # Original fields must still be present
+        assert "nanobot v0.1.0" in content
+        assert "claude-opus-4-5" in content
+        assert "1000 in / 200 out" in content
+        # New field appended
+        assert "duckduckgo" in content
+
+    def test_full_tavily_in_status(self):
+        info = SearchUsageInfo(
+            provider="tavily",
+            supported=True,
+            used=142,
+            limit=1000,
+            remaining=858,
+            reset_date="2026-05-01",
+            search_used=120,
+            extract_used=15,
+            crawl_used=7,
+        )
+        content = build_status_content(**self._BASE_KWARGS, search_usage_text=info.format())
+        assert "142 / 1000" in content
+        assert "858" in content
+        assert "2026-05-01" in content
+        assert "Search: 120" in content