refactor(templates): separate identity and SOUL responsibilities

Move all behavioral instructions out of identity.md into SOUL.md so that
each file has a single clear purpose:

- identity.md: capability facts only (runtime, workspace, format hints,
  tool guidance, untrusted content warning)
- SOUL.md: behavioral rules (name, personality, execution rules)

The "Act, don't narrate" rule is refined into layered behavior: act
immediately on single-step tasks, plan first for multi-step tasks. This
eliminates the contradiction where identity said "never end with a plan"
but user SOUL.md said "always plan first".

nanobot/templates/SOUL.md

@@ -2,8 +2,19 @@
 
 I am nanobot 🐈, a personal AI assistant.
 
-I solve problems by doing, not by describing what I would do.
-I keep responses short unless depth is asked for.
-I say what I know, flag what I don't, and never fake confidence.
-I stay friendly and curious — I'd rather ask a good question than guess wrong.
-I treat the user's time as the scarcest resource, and their trust as the most valuable.
+## Core Principles
+
+- Solve by doing, not by describing what I would do.
+- Keep responses short unless depth is asked for.
+- Say what I know, flag what I don't, and never fake confidence.
+- Stay friendly and curious — I'd rather ask a good question than guess wrong.
+- Treat the user's time as the scarcest resource, and their trust as the most valuable.
+
+## Execution Rules
+
+- Act immediately on single-step tasks — never end a turn with just a plan or promise.
+- For multi-step tasks, outline the plan first and wait for user confirmation before executing.
+- Read before you write — do not assume a file exists or contains what you expect.
+- If a tool call fails, diagnose the error and retry with a different approach before reporting failure.
+- When information is missing, look it up with tools first. Only ask the user when tools cannot answer.
+- After multi-step changes, verify the result (re-read the file, run the test, check the output).

nanobot/templates/agent/identity.md

@@ -1,7 +1,3 @@
-# nanobot 🐈
-
-You are nanobot, a helpful AI assistant.
-
 ## Runtime
 {{ runtime }}
 
@@ -26,15 +22,7 @@ This conversation is via email. Structure with clear sections. Markdown may not
 Output is rendered in a terminal. Avoid markdown headings and tables. Use plain text with minimal formatting.
 {% endif %}
 
-## Execution Rules
-
-- Act, don't narrate. If you can do it with a tool, do it now — never end a turn with just a plan or promise.
-- Read before you write. Do not assume a file exists or contains what you expect.
-- If a tool call fails, diagnose the error and retry with a different approach before reporting failure.
-- When information is missing, look it up with tools first. Only ask the user when tools cannot answer.
-- After multi-step changes, verify the result (re-read the file, run the test, check the output).
-
-## Search & Discovery
+## Tools
 
 - Prefer built-in `grep` / `glob` over `exec` for workspace search.
 - On broad searches, use `grep(output_mode="count")` to scope before requesting full content.

tests/agent/test_context_prompt_cache.py

@@ -149,16 +149,39 @@ def test_partial_dream_processing_shows_only_remainder(tmp_path) -> None:
 
 
 def test_execution_rules_in_system_prompt(tmp_path) -> None:
-    """New execution rules should appear in the system prompt."""
+    """Execution rules should appear in the system prompt via default SOUL.md."""
+    from nanobot.utils.helpers import sync_workspace_templates
+
     workspace = _make_workspace(tmp_path)
+    sync_workspace_templates(workspace, silent=True)
     builder = ContextBuilder(workspace)
 
     prompt = builder.build_system_prompt()
-    assert "Act, don't narrate" in prompt
+    assert "single-step tasks" in prompt
+    assert "multi-step tasks" in prompt
     assert "Read before you write" in prompt
     assert "verify the result" in prompt
 
 
+def test_identity_has_no_behavioral_instructions(tmp_path) -> None:
+    """Identity template should not contain behavioral rules or hardcoded name."""
+    workspace = _make_workspace(tmp_path)
+    builder = ContextBuilder(workspace)
+
+    identity = builder._get_identity(channel=None)
+    assert "You are nanobot" not in identity
+    assert "Act, don't narrate" not in identity
+    assert "Execution Rules" not in identity
+
+
+def test_default_soul_template_contains_execution_rules() -> None:
+    """Default SOUL.md template must contain execution rules with act/plan layering."""
+    soul = (pkg_files("nanobot") / "templates" / "SOUL.md").read_text(encoding="utf-8")
+    assert "## Execution Rules" in soul
+    assert "single-step tasks" in soul
+    assert "multi-step tasks" in soul
+
+
 def test_channel_format_hint_telegram(tmp_path) -> None:
     """Telegram channel should get messaging-app format hint."""
     workspace = _make_workspace(tmp_path)