Grayscale/nanobot
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(goal): /goal command & long-running tasks (long_task)

Browse Source 
* feat(long-task): add LongTaskTool for multi-step agent tasks

Implements a meta-ReAct loop where long-running tasks are broken into
sequential subagent steps, each starting fresh with the original goal
and progress from the previous step. This prevents context drift when
agents work on complex, multi-step tasks.

- Extract build_tool_registry() from SubagentManager for reuse
- Add run_step() for synchronous subagent execution (no bus announcement)
- Add HandoffTool and CompleteTool as signal mechanisms via shared dict
- Add LongTaskTool orchestrator with simplified prompt (8 iterations/step)
- Register LongTaskTool in main agent loop
- Add _extract_handoff_from_messages fallback for robustness

* fix(long-task): add debug logging for step-level observability

* feat(long-task): major overhaul with structured handoffs, validation, and observability

- Structured HandoffState: HandoffTool now accepts files_created,
  files_modified, next_step_hint, and verification fields instead of
  a plain string. Progress is passed between steps as structured data.

- Completion validation round: After complete() is called, a dedicated
  validator step runs to verify the claim against the original goal.
  If validation fails, the task continues rather than returning
  a false completion.

- Dynamic prompt system: 3 Jinja2 templates (step_start, step_middle,
  step_final) selected based on step number. Final steps get tighter
  budget and stronger "wrap up" guidance.

- Automatic file change tracking: Extracts write_file/edit_file events
  from tool_events and injects them into the next step's context if
  the subagent forgot to report them explicitly.

- Budget tracking & adaptive strategy: Cumulative token usage is tracked
  across steps. Per-step tool budget drops from 8 to 4 in the last
  two steps to force handoff/completion.

- Crash retry with graceful degradation: A step that crashes is retried
  once. Persistent crashes terminate the task and return partial progress.

- Full observability hooks for future WebUI integration:
  - set_hooks() with on_step_start, on_step_complete, on_handoff,
    on_validation_started, on_validation_passed, on_validation_failed,
    on_task_complete, on_task_error, and catch-all on_event.
  - Readable state properties: current_step, total_steps, status,
    last_handoff, cumulative_usage, goal.
  - inject_correction() allows external code to send user corrections
    that are injected into the next step's prompt.

- run_step() accepts optional max_iterations for dynamic budget control.

All 27 long-task tests and 11 subagent tests pass.

* test(long-task): add boundary tests and fix race conditions

- Add 7 edge-case tests: validation crash resilience, hook exception safety, mid-run correction injection, FIFO correction ordering, explicit file changes overriding auto-detection, final budget for max_steps=1, and dynamic budget switching boundaries

- Fix assertion in test_long_task_completes_after_multiple_handoffs to match exact prompt format

- Remove asyncio timing hack from test_state_exposure

- Add asyncio.sleep(0) yield in test_inject_correction_during_execution to prevent race between signal injection and step continuation

- All 34 tests passing

* fix(long-task): address code review findings

- Declare _scopes = {"core"} explicitly to prevent recursive nesting in subagent scope
- Document fragile coupling in _extract_file_changes: path extraction depends on
  write_file/edit_file detail format; add debug log for unexpected formats
- Align final-template threshold (max_steps - 2) with budget switch threshold
- Eliminate hasattr(self, "_state") in _reset_state by initializing in __init__

* fix(long-task): honor final signal and file tracking

Co-authored-by: Cursor <cursoragent@cursor.com>

* feat(long-task): improve prompt structure and agent contract

- Expand LongTaskTool.description to instruct parent agent on goal
  construction, return value semantics, and how to handle results.
- Expand CompleteTool.description to emphasize that the summary IS the
  final answer returned to the parent agent.
- Prefix validated return value with an explicit "final answer" directive
  to stop parent agent from re-running work.
- Redesign step_start.md: Step 1 is now explicitly for exploration,
  planning, and skeleton-building. complete() is discouraged.
- Remove bulky payload debug logging from _emit(); add targeted
  info/warning/error logs at key state transitions instead.
- Add signal_type to HandoffState for cleaner signal detection.

* test(long-task): expect wrapped completion message after validation

Align assertions with LongTaskTool final return shape on main.

Co-authored-by: Cursor <cursoragent@cursor.com>

* feat(webui): turn timing strip, latency, and session-switch restore

- Agent loop: publish goal_status run/idle for WebSocket turns; attach
  wall-clock latency_ms on turn_end and persisted assistant metadata.
- WebSocket channel: forward goal_status and latency fields to clients.
- NanobotClient: track goal_status started_at per chat without requiring
  onChat; useNanobotStream restores run strip when returning to a chat.
- Thread UI: composer/shell viewport hooks for run duration and latency;
  format helpers and i18n strings.
- MessageBubble: drop trailing StreamCursor (layout artifact vs block markdown).
- Builtin / tests: model command coverage, websocket and loop tests.

Covers multi-session UX and round-trip timing visibility for the WebUI.

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix: keep message-tool file attachments after canonical history hydrate

- MessageTool records per-turn media paths delivered to the active chat.
- nanobot.utils.session_attachments stages out-of-media-root files and
  merges into the last assistant message before save (loop stays a thin call).
- WebUI MediaCell: use a signed URL as a real download link when present.

Fixes attachments flashing then vanishing on turn_end when paths lived
outside get_media_dir (e.g. workspace files).

Co-authored-by: Cursor <cursoragent@cursor.com>

* feat(webui): agent activity cluster, stable keys, LTR sheen labels

- Group reasoning and tool traces in AgentActivityCluster with i18n summaries
- Stabilize React list keys for activity clusters (first message id anchor)
- Replace background-clip shimmer with overlay sheen for streaming labels
- ThreadMessages/MessageList integration and locale strings

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix(webui): render assistant reasoning with Markdown + deferred stream

- Use MarkdownText for ReasoningBubble body (same GFM/KaTeX path as replies)
- Apply muted/italic prose tokens so thinking stays visually subordinate
- useDeferredValue while reasoningStreaming to ease parser work during deltas
- Preload markdown chunk when trace opens; add regression test with preloaded renderer

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix(webui): default-collapse agent activity cluster while Working

Outer fold no longer auto-expands during isTurnStreaming; user opens to see traces.
Header sheen and live summary unchanged.

Co-authored-by: Cursor <cursoragent@cursor.com>

* feat(long_task): cumulative run history, file union, and prompt tuning

Inject cross-step summaries and merged file paths into middle/final step
templates so chains do not lose early context. Strip the last run-history
block when it duplicates Previous Progress to save tokens. Add optional
cumulative_prompt_max_chars and cumulative_step_body_max_chars parameters
with clamped defaults.

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix(webui): session switch keeps in-flight thread and replays buffered WS

Save the prior chat message list to the per-chat cache in a layout effect
when chatId changes (before stale writes could corrupt another chat).
Skip one post-switch layout cache tick so we do not snapshot the wrong tab.

Buffer inbound events per chat_id when no onChat subscriber is registered
(e.g. user focused another session) and drain on resubscribe up to a cap,
so streaming deltas are not lost while off-tab.

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix(webui): snap thread scroll to bottom on session open (no smooth glide)

Use scroll-behavior auto on the viewport, instant programmatic scroll when
following new messages and on scrollToBottomSignal. Keep smooth only for
the explicit scroll-to-bottom button.

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix(webui): respect manual scroll-up after opening a session

Track when the user leaves the bottom with a ref and skip ResizeObserver
and deferred bottom snaps until they return or the conversation is reset.
Remove the time-based force-bottom window that overrode atBottom.

Multi-frame scrollToBottom honours the same guard unless force (scroll button).

Co-authored-by: Cursor <cursoragent@cursor.com>

* Publish long_task UI snapshots on outbound metadata

- Add OUTBOUND_META_AGENT_UI (_agent_ui) for channel-agnostic structured state
- LongTaskTool publishes {kind: long_task, data: snapshot} on the bus with _progress
- WebSocket send forwards metadata as agent_ui for WebUI clients
- Tests for bus payload, WS frame, and progress assertions
- Fix loop progress tests: ignore _goal_status in streaming final filter and
  avoid brittle outbound[-1] ordering after goal status idle messages

Co-authored-by: Cursor <cursoragent@cursor.com>

* feat: WebUI long_task activity card and resilient history merge

Add optional ui_summary to the long_task tool for one-line UI labels. Stream
long_task agent_ui into a dedicated message row with timeline, markdown peek,
and a right sheet for details. Merge canonical history after turn_end while
re-inserting long_task rows before the final assistant reply. Collapse
duplicate task_start/step_start steps in the timeline and extend i18n.

Co-authored-by: Cursor <cursoragent@cursor.com>

* refactor: align long_task with thread_goal and drop orchestrator UI

- Persist sustained objectives via session metadata (long_task / complete_goal); no subagent wiring or tool-driven agent_ui payloads.\n- Remove WebUI long-task activity UI, types, and translations; history merge preserves trace replay only, with legacy long_task rows normalized to traces.\n- Drop long_task prompt templates and get_long_task_run_dir; add webui thread disk helper for gateway persistence tests.

Co-authored-by: Cursor <cursoragent@cursor.com>

* feat(agent): thread goal runtime context, tools, and skill

- Add thread_goal_state helper and mirror active objectives into Runtime Context
- Wire loop/context/memory/events as needed for goal metadata in turns
- Expand long_task / complete_goal semantics (pivot/cancel/honest recap)
- Add always-on thread-goal SKILL.md; align /goal command prompt
- Tests for context builder and thread goal state
- Remove unused webui ChatPane component

Co-authored-by: Cursor <cursoragent@cursor.com>

* feat(thread-goal): add websocket snapshot helper and publish goal updates from long_task

Introduce thread_goal_ws_blob for bounded JSON snapshots, attach snapshots to
websocket turn_end metadata in AgentLoop, and let long_task fan-out dedicated
thread_goal frames on the websocket channel after persisting session metadata.

Co-authored-by: Cursor <cursoragent@cursor.com>

* feat(channels): websocket thread_goal frames, turn_end replay, and session API scrub for subagent inject

Emit thread_goal events and optional thread_goal on turn_end; scrub persisted
subagent announce blobs on GET /api/sessions/.../messages and shorten session
list previews so WebUI does not surface full Task/Summarize scaffolding.

Co-authored-by: Cursor <cursoragent@cursor.com>

* feat(webui): merge ephemeral traces per user turn when reconciling canonical history

Preserve disk/live trace rows inside the matching user–assistant segment instead
of stacking every trace before the final assistant reply (fixes inflated tool
counts after refresh or session switch).

Co-authored-by: Cursor <cursoragent@cursor.com>

* feat(webui): show assistant reply copy only on the last slice before the next user turn

Avoid duplicate copy affordances on intermediate assistant bubbles that precede
more agent activity in the same turn (tools or further assistant text).

Co-authored-by: Cursor <cursoragent@cursor.com>

* feat(webui): thread_goal stream plumbing, composer goal strip, sky glow, and client-side subagent scrub projection

Track thread_goal and turn_goal snapshots in NanobotClient, hydrate React state
from thread_goal frames and turn_end, surface objective/elapsed in the composer,
add breathing sky halo CSS while goals are active, mirror server scrub logic on
history hydration and webui_thread snapshots, and extend tests/client mocks.

Co-authored-by: Cursor <cursoragent@cursor.com>

* feat(channels): add Slack Socket Mode connect timeout with actionable timeout errors

Abort hung websockets.connect handshakes after a bounded wait, log REST-vs-WSS
guidance, surface RuntimeError to channel startup, and log successful WSS setup.

Co-authored-by: Cursor <cursoragent@cursor.com>

* webui: expand thread goal in composer bottom sheet

Add ChevronUp control on the run/goal strip that opens a bottom Sheet
with full ui_summary and objective. Inline preview logic in RunElapsedStrip,
add i18n strings across locales, and a composer unit test.

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix(webui): widen dedupeToolCallsForUi input for session API typing

fetchSessionMessages types tool_calls as unknown; accept unknown so tsc
build passes when passing message.tool_calls through.

Co-authored-by: Cursor <cursoragent@cursor.com>

* refactor(agent): extract WebSocket turn run status to webui_turn_helpers

* refactor(skills): rename thread-goal to long-task and document idempotent goals

* feat(skills): rename sustained-goal skill to long-goal and tighten long_task guidance

* chore: remove unused subagent/context/router helpers

* feat(session): rename sustained goal to goal_state and align WS/WebUI

- Move helpers from agent/thread_goal_state to session/goal_state:
  GOAL_STATE_KEY, goal_state_runtime_lines, goal_state_ws_blob, parse_goal_state.
- Session metadata now uses "goal_state"; still read legacy "thread_goal";
  long_task writes drop the legacy key after save.
- WebSocket: event/field goal_state, _goal_state_sync; turn_end carries goal_state;
  accept legacy _thread_goal_sync/thread_goal inbound metadata for dispatch.
- WebUI: GoalStateWsPayload, goalState hook/client props, i18n keys goalState*.
- Runtime Context copy uses "Goal (active):" instead of "Thread goal".

* feat(agent): stream Anthropic thinking deltas and fix stream idle timeout

* refactor(webui): transcript jsonl as sole timeline source

* fix(agent): reject mismatched WS message chat_id and stream reasoning deltas

* feat(webui): hydrate sustained goal and run timer after websocket subscribe

* chore(webui,websocket): remove unused fetch helpers and legacy thread_goal WS paths

* Raise default max_tokens and context window in agent schema.

Align AgentDefaults and ModelPresetConfig with typical Claude-scale usage
(32k completion budget, 256k context window) and update migration tests.

Co-authored-by: Cursor <cursoragent@cursor.com>

* feat(gateway): bootstrap prefers in-memory model; clarify websocket naming

* fix(websocket): websocket _handle_message passes is_dm; refresh /status test expectations

---------

Co-authored-by: chengyongru <2755839590@qq.com>
Co-authored-by: chengyongru <chengyongru.ai@gmail.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
Xubin Ren
2026-05-16 01:14:11 +08:00
committed by GitHub
co-authored by Cursor chengyongru chengyongru
parent 2d17a095dc
commit 1c2ea1aad2
99 changed files with 4886 additions and 848 deletions
Download Patch File Download Diff File Expand all files Collapse all files
nanobot/utils/session_attachments.py
+74
View File
@@ -0,0 +1,74 @@
"""Session replay: ensure assistant ``media`` paths are under the media root.
WebUI history signing (``/api/.../messages``) only works for files inside
``get_media_dir``. Tool-driven attachments may live in the workspace; stage
copies into the websocket media bucket before persisting message JSON.
"""
from __future__ import annotations
import shutil
import uuid
from pathlib import Path
from typing import Any
from loguru import logger
from nanobot.config.paths import get_media_dir
from nanobot.utils.helpers import safe_filename
def stage_media_paths_for_session_replay(paths: list[str]) -> list[str]:
"""Keep local files only; copy anything outside the media root into ``media/websocket``."""
root = get_media_dir().resolve()
out: list[str] = []
seen: set[str] = set()
for raw in paths:
if not isinstance(raw, str) or not raw.strip():
continue
if raw.startswith(("http://", "https://")):
continue
try:
p = Path(raw).expanduser().resolve()
except OSError:
continue
if not p.is_file():
continue
try:
p.relative_to(root)
key = str(p)
except ValueError:
try:
media_dir = get_media_dir("websocket")
staged = media_dir / f"{uuid.uuid4().hex[:12]}-{safe_filename(p.name) or 'attachment'}"
shutil.copyfile(p, staged)
key = str(staged.resolve())
except OSError as exc:
logger.warning("failed to stage session media from {}: {}", raw, exc)
continue
if key not in seen:
out.append(key)
seen.add(key)
return out
def merge_turn_media_into_last_assistant(
all_messages: list[dict[str, Any]],
generated_image_paths: list[str],
extra_attachment_paths: list[str],
) -> None:
"""Attach staged paths to the last assistant row in *all_messages* (in-place)."""
merged = list(
dict.fromkeys(
[
*stage_media_paths_for_session_replay(generated_image_paths),
*stage_media_paths_for_session_replay(extra_attachment_paths),
]
)
)
last = all_messages[-1] if all_messages else None
if not merged or not last or last.get("role") != "assistant":
return
existing = last.get("media")
base = existing if isinstance(existing, list) else []
last["media"] = list(dict.fromkeys([*base, *merged]))
nanobot/utils/subagent_channel_display.py
+59
View File
@@ -0,0 +1,59 @@
"""Strip internal subagent inject scaffolding for human-facing channel surfaces.
Persisted subagent announcements mirror ``agent/subagent_announce.md``: header,
full ``Task:`` assignment (model context), ``Result:``, and a trailing model-only
``Summarize…`` instruction. External channels (embedded WebUI, session previews)
should show only the header plus a truncated result body."""
from __future__ import annotations
from typing import Any
# Cap Result section length so WebSocket session replay stays readable; full text
# remains on disk for LLM replay (we only mutate outgoing API copies in websocket).
_SUBAGENT_CHANNEL_RESULT_MAX_CHARS = 800
def scrub_subagent_announce_body(content: str) -> str:
"""Return channel-safe text derived from a full subagent announce blob."""
stripped = content.replace("\r\n", "\n").strip()
lines = stripped.splitlines()
header = ""
if lines and lines[0].startswith("[Subagent"):
header = lines[0].strip()
lower = stripped.lower()
key = "\nresult:\n"
ri = lower.find(key)
if ri == -1:
key = "\nresult:"
ri = lower.find(key)
if ri == -1:
return header if header else stripped
after = stripped[ri + len(key) :].lstrip()
summ_marker = "summarize this naturally"
si = after.lower().find(summ_marker)
if si != -1:
after = after[:si].rstrip()
body = after.strip()
limit = _SUBAGENT_CHANNEL_RESULT_MAX_CHARS
if limit and len(body) > limit:
body = body[: limit - 1].rstrip() + "…"
if header and body:
return f"{header}\n\n{body}"
return header or body or stripped
def scrub_subagent_messages_for_channel(messages: list[dict[str, Any]]) -> None:
"""Mutate message dicts in place when they carry ``subagent_result`` inject."""
for msg in messages:
if not isinstance(msg, dict):
continue
if msg.get("injected_event") != "subagent_result":
continue
raw = msg.get("content")
if not isinstance(raw, str) or not raw.strip():
continue
msg["content"] = scrub_subagent_announce_body(raw)
nanobot/utils/webui_thread_disk.py
+31
View File
@@ -0,0 +1,31 @@
"""Legacy WebUI JSON snapshot path helpers (JSON file); transcripts use webui_transcript."""
from __future__ import annotations
from pathlib import Path
from loguru import logger
from nanobot.config.paths import get_webui_dir
from nanobot.session.manager import SessionManager
from nanobot.utils.webui_transcript import delete_webui_transcript
def webui_thread_file_path(session_key: str) -> Path:
stem = SessionManager.safe_key(session_key)
return get_webui_dir() / f"{stem}.json"
def delete_webui_thread(session_key: str) -> bool:
"""Remove legacy WebUI JSON snapshot and append-only transcript for *session_key*."""
removed = False
path = webui_thread_file_path(session_key)
if path.is_file():
try:
path.unlink()
removed = True
except OSError as e:
logger.warning("Failed to delete webui thread file {}: {}", path, e)
if delete_webui_transcript(session_key):
removed = True
return removed
nanobot/utils/webui_transcript.py
+423
View File
@@ -0,0 +1,423 @@
"""Append-only WebUI display transcript (JSONL), separate from agent session."""
from __future__ import annotations
import json
import os
import time
import uuid
from pathlib import Path
from typing import Any, Callable
from loguru import logger
from nanobot.config.paths import get_webui_dir
from nanobot.session.manager import SessionManager
WEBUI_TRANSCRIPT_SCHEMA_VERSION = 3
_MAX_TRANSCRIPT_FILE_BYTES = 8 * 1024 * 1024
def webui_transcript_path(session_key: str) -> Path:
stem = SessionManager.safe_key(session_key)
return get_webui_dir() / f"{stem}.jsonl"
def read_transcript_lines(session_key: str) -> list[dict[str, Any]]:
path = webui_transcript_path(session_key)
if not path.is_file():
return []
size = path.stat().st_size
if size > _MAX_TRANSCRIPT_FILE_BYTES:
logger.warning("webui transcript too large, skipping: {}", path)
return []
lines_out: list[dict[str, Any]] = []
try:
with open(path, encoding="utf-8") as f:
for line_no, line in enumerate(f, start=1):
line = line.strip()
if not line:
continue
try:
obj = json.loads(line)
except json.JSONDecodeError:
logger.warning("bad jsonl at {} line {}", path, line_no)
continue
if isinstance(obj, dict):
lines_out.append(obj)
except OSError as e:
logger.warning("read transcript failed {}: {}", path, e)
return []
return lines_out
def append_transcript_object(session_key: str, obj: dict[str, Any]) -> None:
raw = json.dumps(obj, ensure_ascii=False, separators=(",", ":"))
if len(raw.encode("utf-8")) > _MAX_TRANSCRIPT_FILE_BYTES:
msg = "webui transcript line too large"
raise ValueError(msg)
path = webui_transcript_path(session_key)
path.parent.mkdir(parents=True, exist_ok=True)
line = raw + "\n"
with open(path, "a", encoding="utf-8") as f:
f.write(line)
f.flush()
os.fsync(f.fileno())
def delete_webui_transcript(session_key: str) -> bool:
path = webui_transcript_path(session_key)
if not path.is_file():
return False
try:
path.unlink()
return True
except OSError as e:
logger.warning("Failed to delete webui transcript {}: {}", path, e)
return False
def _format_tool_call_trace(call: Any) -> str | None:
if not call or not isinstance(call, dict):
return None
fn = call.get("function")
name = fn.get("name") if isinstance(fn, dict) else None
if not isinstance(name, str) or not name:
raw_name = call.get("name")
name = raw_name if isinstance(raw_name, str) else ""
if not name:
return None
args = (fn.get("arguments") if isinstance(fn, dict) else None) or call.get("arguments")
if isinstance(args, str) and args.strip():
return f"{name}({args})"
if args and isinstance(args, dict):
return f"{name}({json.dumps(args, ensure_ascii=False)})"
return f"{name}()"
def tool_trace_lines_from_events(events: Any) -> list[str]:
if not isinstance(events, list):
return []
lines: list[str] = []
for event in events:
if not event or not isinstance(event, dict):
continue
if event.get("phase") != "start":
continue
t = _format_tool_call_trace(event)
if t:
lines.append(t)
return lines
def replay_transcript_to_ui_messages(
lines: list[dict[str, Any]],
*,
augment_user_media: Callable[[list[str]], list[dict[str, Any]]] | None = None,
) -> list[dict[str, Any]]:
"""Fold JSONL records into ``UIMessage``-shaped dicts for the WebUI.
Mirrors the core fold in ``useNanobotStream.ts`` (delta, reasoning,
message+kind, turn_end). ``augment_user_media`` maps persisted filesystem
paths to ``{url, name?}`` / attachment dicts the client expects.
"""
messages: list[dict[str, Any]] = []
buffer_message_id: str | None = None
buffer_parts: list[str] = []
suppress_until_turn_end = False
_ts_base = int(time.time() * 1000)
def _new_id(prefix: str, idx: int) -> str:
return f"{prefix}-{idx}-{uuid.uuid4().hex[:8]}"
def attach_reasoning_chunk(prev: list[dict[str, Any]], chunk: str, idx: int) -> None:
for i in range(len(prev) - 1, -1, -1):
candidate = prev[i]
if candidate.get("role") == "user":
break
if candidate.get("kind") == "trace":
break
if candidate.get("role") != "assistant":
continue
content = str(candidate.get("content") or "")
has_answer = len(content) > 0
if (
candidate.get("reasoningStreaming")
or candidate.get("reasoning") is not None
or has_answer
or candidate.get("isStreaming")
):
prev[i] = {
**candidate,
"reasoning": (str(candidate.get("reasoning") or "")) + chunk,
"reasoningStreaming": True,
}
return
if not has_answer and candidate.get("isStreaming"):
prev[i] = {**candidate, "reasoning": chunk, "reasoningStreaming": True}
return
break
prev.append(
{
"id": _new_id("as", idx),
"role": "assistant",
"content": "",
"isStreaming": True,
"reasoning": chunk,
"reasoningStreaming": True,
"createdAt": _ts_base + idx,
},
)
def find_active_placeholder(prev: list[dict[str, Any]]) -> str | None:
last = prev[-1] if prev else None
if not last:
return None
if last.get("role") != "assistant" or last.get("kind") == "trace":
return None
if str(last.get("content") or ""):
return None
if not last.get("isStreaming"):
return None
return str(last.get("id"))
def close_reasoning(prev: list[dict[str, Any]]) -> None:
for i in range(len(prev) - 1, -1, -1):
if prev[i].get("reasoningStreaming"):
prev[i] = {**prev[i], "reasoningStreaming": False}
return
def is_reasoning_only_placeholder(m: dict[str, Any]) -> bool:
return (
m.get("role") == "assistant"
and m.get("kind") != "trace"
and not str(m.get("content") or "").strip()
and bool(m.get("reasoning"))
and not m.get("reasoningStreaming")
and not m.get("media")
)
def is_tool_trace_at(index: int) -> bool:
m = messages[index] if 0 <= index < len(messages) else None
return bool(m and m.get("kind") == "trace")
def prune_reasoning_only() -> None:
nonlocal messages
kept: list[dict[str, Any]] = []
for i, m in enumerate(messages):
if is_reasoning_only_placeholder(m) and not is_tool_trace_at(i + 1):
continue
kept.append(m)
messages = kept
def stamp_latency(latency_ms: int) -> None:
for i in range(len(messages) - 1, -1, -1):
if messages[i].get("role") == "assistant" and messages[i].get("kind") != "trace":
messages[i] = {
**messages[i],
"latencyMs": latency_ms,
"isStreaming": False,
}
return
def absorb_complete(extra: dict[str, Any], idx: int) -> None:
last = messages[-1] if messages else None
if last and is_reasoning_only_placeholder(last):
messages[-1] = {
**last,
**extra,
"isStreaming": False,
"reasoningStreaming": False,
}
else:
messages.append(
{
"id": _new_id("as", idx),
"role": "assistant",
"createdAt": _ts_base + idx,
**extra,
},
)
for idx, rec in enumerate(lines):
ev = rec.get("event")
if ev == "user":
text = rec.get("text")
text_s = text if isinstance(text, str) else ""
media_paths = rec.get("media_paths")
paths: list[str] = []
if isinstance(media_paths, list):
paths = [str(p) for p in media_paths if p]
media_att: list[dict[str, Any]] | None = None
if paths and augment_user_media is not None:
media_att = augment_user_media(paths)
row: dict[str, Any] = {
"id": _new_id("u", idx),
"role": "user",
"content": text_s,
"createdAt": _ts_base + idx,
}
if media_att:
row["media"] = media_att
if all(m.get("kind") == "image" for m in media_att):
row["images"] = [{"url": m.get("url"), "name": m.get("name")} for m in media_att]
messages.append(row)
continue
if ev == "delta":
if suppress_until_turn_end:
continue
chunk = rec.get("text")
if not isinstance(chunk, str):
continue
adopted = find_active_placeholder(messages) if buffer_message_id is None else None
if buffer_message_id is None:
if adopted:
buffer_message_id = adopted
else:
buffer_message_id = _new_id("buf", idx)
messages.append(
{
"id": buffer_message_id,
"role": "assistant",
"content": "",
"isStreaming": True,
"createdAt": _ts_base + idx,
},
)
buffer_parts.append(chunk)
combined = "".join(buffer_parts)
for i, m in enumerate(messages):
if m.get("id") == buffer_message_id:
messages[i] = {**m, "content": combined, "isStreaming": True}
break
continue
if ev == "stream_end":
if suppress_until_turn_end:
buffer_message_id = None
buffer_parts = []
continue
buffer_message_id = None
buffer_parts = []
continue
if ev == "reasoning_delta":
if suppress_until_turn_end:
continue
chunk = rec.get("text")
if not isinstance(chunk, str) or not chunk:
continue
attach_reasoning_chunk(messages, chunk, idx)
continue
if ev == "reasoning_end":
if suppress_until_turn_end:
continue
close_reasoning(messages)
continue
if ev == "message":
if suppress_until_turn_end and rec.get("kind") in (
"tool_hint",
"progress",
"reasoning",
):
continue
kind = rec.get("kind")
if kind == "reasoning":
line = rec.get("text")
if not isinstance(line, str) or not line:
continue
attach_reasoning_chunk(messages, line, idx)
close_reasoning(messages)
continue
if kind in ("tool_hint", "progress"):
structured = tool_trace_lines_from_events(rec.get("tool_events"))
text = rec.get("text")
trace_lines = structured if structured else ([text] if isinstance(text, str) and text else [])
if not trace_lines:
continue
last = messages[-1] if messages else None
if last and last.get("kind") == "trace" and not last.get("isStreaming"):
prev_traces = list(last.get("traces") or [last.get("content")])
merged_traces = prev_traces + trace_lines
messages[-1] = {
**last,
"traces": merged_traces,
"content": trace_lines[-1],
}
else:
messages.append(
{
"id": _new_id("tr", idx),
"role": "tool",
"kind": "trace",
"content": trace_lines[-1],
"traces": trace_lines,
"createdAt": _ts_base + idx,
},
)
continue
buffer_message_id = None
buffer_parts = []
text = rec.get("text")
content_s = text if isinstance(text, str) else ""
media_urls = rec.get("media_urls")
media: list[dict[str, Any]] = []
if isinstance(media_urls, list):
for m in media_urls:
if isinstance(m, dict) and m.get("url"):
media.append(
{
"kind": "image",
"url": str(m["url"]),
"name": str(m.get("name") or ""),
},
)
extra: dict[str, Any] = {"content": content_s}
if media:
extra["media"] = media
lat = rec.get("latency_ms")
if isinstance(lat, (int, float)) and lat >= 0:
extra["latencyMs"] = int(lat)
absorb_complete(extra, idx)
if media:
suppress_until_turn_end = True
continue
if ev == "turn_end":
suppress_until_turn_end = False
for i, m in enumerate(messages):
if m.get("isStreaming"):
messages[i] = {**m, "isStreaming": False}
prune_reasoning_only()
lat = rec.get("latency_ms")
if isinstance(lat, (int, float)) and lat >= 0:
stamp_latency(int(lat))
buffer_message_id = None
buffer_parts = []
continue
for m in messages:
m.pop("isStreaming", None)
m.pop("reasoningStreaming", None)
return messages
def build_webui_thread_response(
session_key: str,
*,
augment_user_media: Callable[[list[str]], list[dict[str, Any]]] | None = None,
) -> dict[str, Any] | None:
"""Return a payload compatible with ``WebuiThreadPersistedPayload``."""
lines = read_transcript_lines(session_key)
if not lines:
return None
msgs = replay_transcript_to_ui_messages(lines, augment_user_media=augment_user_media)
return {
"schemaVersion": WEBUI_TRANSCRIPT_SCHEMA_VERSION,
"sessionKey": session_key,
"messages": msgs,
}
nanobot/utils/webui_turn_helpers.py
+48
View File
@@ -0,0 +1,48 @@
"""Outbound helpers for the WebSocket/WebUI wire contract.
AgentLoop uses these without importing a concrete channel plugin; only
``channel == "websocket"`` messages are affected.
"""
from __future__ import annotations
import time
from typing import Any
from nanobot.bus.events import InboundMessage, OutboundMessage
from nanobot.bus.queue import MessageBus
# Wall-clock turn start per ``chat_id`` (websocket only). Survives browser refresh while the
# gateway process stays up; cleared on idle/stop and implicitly dropped on restart.
_WEBSOCKET_TURN_WALL_STARTED_AT: dict[str, float] = {}
def websocket_turn_wall_started_at(chat_id: str) -> float | None:
"""Return ``time.time()`` when the active user turn began, if still running."""
return _WEBSOCKET_TURN_WALL_STARTED_AT.get(chat_id)
async def publish_turn_run_status(bus: MessageBus, msg: InboundMessage, status: str) -> None:
"""Notify WebSocket clients while a user turn is executing (timing strip)."""
if msg.channel != "websocket":
return
cid = str(msg.chat_id)
meta: dict[str, Any] = {
**dict(msg.metadata or {}),
"_goal_status": True,
"goal_status": status,
}
if status == "running":
t0 = time.time()
meta["started_at"] = t0
_WEBSOCKET_TURN_WALL_STARTED_AT[cid] = t0
else:
_WEBSOCKET_TURN_WALL_STARTED_AT.pop(cid, None)
await bus.publish_outbound(
OutboundMessage(
channel=msg.channel,
chat_id=cid,
content="",
metadata=meta,
),
)