Files
odysseus/src/agent_tools.py

141 lines
5.1 KiB
Python
Raw Normal View History

2026-05-31 23:58:26 +09:00
"""
agent_tools.py — Facade module.
Re-exports tool parsing, schemas, execution, and implementations
for backward compatibility. All importers continue to work unchanged.
Sub-modules:
- tool_parsing.py: regex patterns, parse/strip functions
- tool_schemas.py: FUNCTION_TOOL_SCHEMAS, function_call_to_tool_block
- tool_execution.py: execute_tool_block, format_tool_result, MCP helpers
- tool_implementations.py: all do_* tool functions
"""
import logging
from collections import namedtuple
logger = logging.getLogger(__name__)
# ---------------------------------------------------------------------------
# Constants (kept here — sub-modules import from here)
# ---------------------------------------------------------------------------
cookbook agent debug loop: persistent log files, auto-adopt orphan tmux, Codex/Claude skill parity Three converging fixes so the chat agent + external Codex/Claude skills can actually debug a crashed serve instead of staring at a post-crash neofetch banner: * Serves now `tee` to /tmp/odysseus-tmux/SESSION.log on the host running them. Runner saves fds 3/4 before the tee and restores them right before `exec ${SHELL}`, so the post-crash interactive zsh banner does NOT pollute the log file. * `tail_serve_output` (chat agent) and `/api/codex/cookbook/output/{sid}` (Codex+Claude skills) both prefer the persistent log file over the tmux pane. Pane is fallback for sessions predating the tee runner. Default tail bumped 150 -> 400. * `list_served_models` "recent log" snippet seeks to the Traceback line instead of showing the last 6 lines (which was always the bash prompt). Cookbook auto-adoption sweep on `/api/cookbook/tasks/status`: every 20s (rate-limited) the cookbook SSHes each configured server, finds `serve-*` / `cookbook-*` tmux sessions running an actual model process (vllm/python/llama-server/etc., filtered via `pane_current_command`), and writes them into state.tasks. So when the agent falls back to raw ssh+tmux, the session appears in the Cookbook UI on the next poll. `serve_model` error path now reads `data["detail"]` in addition to `data["error"]` so the FastAPI HTTPException message ("Invalid characters in cmd") actually reaches the agent instead of being swallowed as a generic "Serve failed". Tool description updated to warn against `cd …`/`source …`/`&&` prefixes. Intent-without-action supervisor in agent_loop: when the model writes "Let me tail the output" / "I'll check the logs" / "Let me investigate" and ends the turn without emitting a tool call, the loop injects a sharp system nudge ("You said you would X — DO IT NOW") and continues. Capped at 2 nudges per chat so a model that genuinely cannot use the tool does not pin the loop. Codex/Claude skill parity: adds `/cookbook/cached`, `/cookbook/presets`, `/cookbook/preset/{name}`, `/cookbook/adopt` so external agents have the same surface as the chat agent. SKILL.md docs + odysseus_api.py wrapper updated for both bundles. `adopt_served_model` promoted to the always-on tool set so the agent has a documented fallback when serve_model rejects a cmd. Also various cookbook UI tweaks accumulated alongside the above (cookbook.js, cookbookRunning.js, cookbookServe.js, cookbook-diagnosis.js, settings.js, style.css).
2026-06-04 23:27:18 +09:00
MAX_AGENT_ROUNDS = 50
2026-05-31 23:58:26 +09:00
SHELL_TIMEOUT = 60
PYTHON_TIMEOUT = 30
MAX_OUTPUT_CHARS = 10_000
MAX_READ_CHARS = 20_000
# Tool types that trigger execution
TOOL_TAGS = {"bash", "python", "web_search", "web_fetch", "read_file", "write_file", "edit_file",
"grep", "glob", "ls",
2026-05-31 23:58:26 +09:00
"create_document", "update_document", "edit_document",
"search_chats",
"chat_with_model", "create_session", "list_sessions",
"send_to_session",
"pipeline",
"manage_session", "manage_memory", "list_models",
feat: Add plan mode to the chat agent (#638) * feat: Add plan mode to the chat agent Adds a plan mode: the agent investigates read-only, proposes a checklist, and waits for approval before changing anything. On approval it runs with full tools and checks items off as it goes. Enforcement reuses the existing disabled_tools gate. Includes a slash command: `/plan [on|off]` (and `/toggle plan`) to flip the plan toggle from the chat input. - src/tool_security.py, src/mcp_manager.py: read-only allowlist (tools + MCP). - src/agent_loop.py, routes/chat_routes.py: union the disabled set, prepend the plan directive, force agent mode. - static/: plan toggle pill, Approve & Run, dockable plan window, task-list checkboxes, and the /plan slash command. - tests/test_plan_mode.py. * Plan mode: persistent re-referenceable plan + agent write-back Three improvements so a long plan survives a weak model and stays in reach: 1. Re-reference the plan (out-of-context fix). On the execution turn the frontend sends the approved checklist back (`approved_plan`); the backend pins it as a top-of-context `## ACTIVE PLAN` system note (kept by the context trimmer), so the agent can always re-read the plan instead of losing the thread on a long run. New `build_active_plan_note()` (unit-tested). 2. Re-open / dock the plan anytime. The plan checklist is stored per-session (localStorage). When a plan exists, the plan-mode button opens a small menu ("Show plan" / "Plan mode: On/Off") that re-opens the side-dockable plan window — so it can stay docked while the agent works. The window live-refreshes as the plan changes. 3. Agent write-back: new `update_plan` tool. The agent calls it to tick steps `- [x]` after finishing them, or to revise steps when the user asks. Marker tool (no I/O) → `plan_update` SSE event → the stored plan + docked window update live. The ACTIVE PLAN note instructs the agent to use it. Backend: src/agent_loop.py (param + pin + note builder + emit + prompt blurb), src/tool_execution.py (update_plan handler), routes/chat_routes.py (parse `approved_plan`, relay `plan_update`), registration in tool_schemas / agent_tools / tool_index (always-available, not admin-gated). Frontend: static/js/chat.js (plan store, send `approved_plan`, handle `plan_update`, capture restated checklists), static/app.js (plan-button menu), static/js/planWindow.js (`isPlanWindowOpen`), static/js/storage.js (PLAN key). Tests: tests/test_plan_mode.py (plan-note), tests/test_update_plan_tool.py. * Plan mode: drop bash/python, rely on read-only discovery tools Shell can mutate (write files, hit the network) and can't be constrained to read-only at the tool layer, so plan mode no longer relies on a prompt to keep it well-behaved — bash/python are removed from the read-only allowlist and added to the fail-closed block set. Discovery is covered by the dedicated read-only tools (read_file, grep, glob, ls) instead. Rewrites the plan-mode directive to state shell is disabled and lists the available read-only tools positively. Addresses review feedback on #638. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * Comment: note _MCP_READONLY_VERBS are prefixes not whole words Clarifies that entries like "summar" are intentional stems matched via startswith (covers summarise/summarize/summary), not typos. Addresses review feedback on #638. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * Plan mode: clarify why gating inverts the allowlist into a denylist Rename _PLAN_MODE_FALLBACK_BLOCK -> _PLAN_MODE_KNOWN_MUTATORS and rewrite the comments. The tool gate is a denylist (disabled_tools); plan mode's policy is an allowlist, so it returns the inverse (all known tool names minus the allowlist). The static mutator set is a backstop for the schema-derived name list, which misses XML-only tools and can fail to import. Addresses review feedback on #638. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * Plan mode: stop hardcoding the read-only tool list in the directive The model is already shown its available (read-only) tools by _assemble_prompt, which removes every disabled tool. Enumerating them again in the directive only duplicated that list and would drift as tools change. Point at the tools listed below instead. Addresses review feedback on #638.
2026-06-05 16:32:25 +02:00
"ui_control", "generate_image", "ask_user", "update_plan",
2026-05-31 23:58:26 +09:00
"manage_tasks", "api_call", "ask_teacher", "manage_skills",
"suggest_document",
"manage_endpoints", "manage_mcp", "manage_webhooks",
"manage_tokens", "manage_documents", "manage_settings",
"manage_notes", "manage_calendar",
"resolve_contact", "manage_contact", "list_email_accounts", "send_email", "list_emails",
"read_email", "reply_to_email", "bulk_email", "archive_email",
"delete_email", "mark_email_read",
# Cookbook tools (LLM serving + downloads). Without these
# entries, native function calls to e.g. list_served_models
# are rejected as "Unknown function call" before reaching
# the dispatcher — silent failure for the whole cookbook
# surface.
"download_model", "serve_model",
"list_served_models", "stop_served_model",
"list_downloads", "cancel_download",
"search_hf_models", "list_cached_models",
"list_serve_presets", "serve_preset", "adopt_served_model",
"list_cookbook_servers",
# Other tools the agent reaches for that were also missing.
"edit_image", "trigger_research", "manage_research",
# Generic loopback to any UI-button endpoint (cookbook,
# gallery, email folders, etc.) — agent uses this when
# there's no named tool wrapper for the action.
"app_api"}
ToolBlock = namedtuple("ToolBlock", ["tool_type", "content"])
# ---------------------------------------------------------------------------
# MCP Manager (kept here — used by execution and agent_loop)
# ---------------------------------------------------------------------------
_mcp_manager = None
def set_mcp_manager(manager):
"""Set the global MCP manager instance."""
global _mcp_manager
_mcp_manager = manager
def get_mcp_manager():
"""Get the global MCP manager instance."""
return _mcp_manager
# ---------------------------------------------------------------------------
# Helpers (kept here — used by sub-modules)
# ---------------------------------------------------------------------------
def _truncate(text: str, limit: int = MAX_OUTPUT_CHARS) -> str:
# Callers treat the result as text, so always return a string: coerce a
# non-string (None -> "", otherwise str(...)) instead of returning it raw,
# which would just move the crash downstream.
if not isinstance(text, str):
text = "" if text is None else str(text)
2026-05-31 23:58:26 +09:00
if len(text) > limit:
return text[:limit] + f"\n... (truncated, {len(text)} chars total)"
return text
# ---------------------------------------------------------------------------
# Re-exports from sub-modules
# ---------------------------------------------------------------------------
# Parsing
from src.tool_parsing import ( # noqa: E402, F401
parse_tool_blocks,
strip_tool_blocks,
_TOOL_NAME_MAP,
_TOOL_BLOCK_RE,
_TOOL_CALL_RE,
_XML_TOOL_CALL_RE,
_XML_INVOKE_RE,
_XML_PARAM_RE,
)
# Schemas
from src.tool_schemas import ( # noqa: E402, F401
FUNCTION_TOOL_SCHEMAS,
function_call_to_tool_block,
)
# Execution
from src.tool_execution import ( # noqa: E402, F401
execute_tool_block,
format_tool_result,
)
# Implementations
from src.tool_implementations import ( # noqa: E402, F401
set_active_document,
set_active_model,
get_active_document,
do_create_document,
do_update_document,
do_edit_document,
do_suggest_document,
do_search_chats,
do_manage_skills,
do_manage_tasks,
do_manage_endpoints,
do_manage_mcp,
do_manage_webhooks,
do_manage_tokens,
do_manage_documents,
do_manage_settings,
do_api_call,
)