FoundationAgents/OpenManus · warning · ToolError

Unrecognized command: {command}. Allowed commands are: creat

Error message

Unrecognized command: {command}. Allowed commands are: create, update, list, get, set_active, mark_step, delete

What it means

Raised by the planning tool's execute() dispatch (app/tool/planning.py:116) when the command string matches none of the seven supported commands (create, update, list, get, set_active, mark_step, delete). It is the fallback branch of an if/elif chain; note that execute() also declares command as a Literal type, so well-typed callers get static errors first — this guard catches untyped/dynamic callers (e.g. LLM-issued tool calls).

Source

Thrown at app/tool/planning.py:116

        - step_notes: Additional notes for a step (used with mark_step command)
        """

        if command == "create":
            return self._create_plan(plan_id, title, steps)
        elif command == "update":
            return self._update_plan(plan_id, title, steps)
        elif command == "list":
            return self._list_plans()
        elif command == "get":
            return self._get_plan(plan_id)
        elif command == "set_active":
            return self._set_active_plan(plan_id)
        elif command == "mark_step":
            return self._mark_step(plan_id, step_index, step_status, step_notes)
        elif command == "delete":
            return self._delete_plan(plan_id)
        else:
            raise ToolError(
                f"Unrecognized command: {command}. Allowed commands are: create, update, list, get, set_active, mark_step, delete"
            )

    def _create_plan(
        self, plan_id: Optional[str], title: Optional[str], steps: Optional[List[str]]
    ) -> ToolResult:
        """Create a new plan with the given ID, title, and steps."""
        if not plan_id:
            raise ToolError("Parameter `plan_id` is required for command: create")

        if plan_id in self.plans:
            raise ToolError(
                f"A plan with ID '{plan_id}' already exists. Use 'update' to modify existing plans."
            )

        if not title:
            raise ToolError("Parameter `title` is required for command: create")

View on GitHub (pinned to 52a13f2a57)

Solutions

1. Use one of the exact allowed strings: create, update, list, get, set_active, mark_step, delete (lowercase, underscore-separated).
2. Normalize and validate model-issued commands before dispatch: strip().lower() and check membership in the allowed set, with a corrective message back to the model.
3. After tool upgrades, grep prompts/schemas for stale command names and update them.

Example fix

// before
await planning.execute(command='delete_plan', plan_id='p1')

// after
ALLOWED = {'create','update','list','get','set_active','mark_step','delete'}
cmd = raw_cmd.strip().lower()
if cmd not in ALLOWED:
    return f"Unknown command {raw_cmd!r}; use one of {sorted(ALLOWED)}"
await planning.execute(command=cmd, plan_id='p1')

ALLOWED = ('create', 'update', 'list', 'get', 'set_active', 'mark_step', 'delete')
cmd = raw.strip().lower()
if cmd not in ALLOWED:
    return f'Unknown command {raw!r}; allowed: {ALLOWED}'
await planning.execute(command=cmd, **params)

from typing import Literal
PlanningCommand = Literal['create','update','list','get','set_active','mark_step','delete']
def is_planning_command(c: str) -> TypeGuard[PlanningCommand]:
    return c in {'create','update','list','get','set_active','mark_step','delete'}

try:
    await planning.execute(command=cmd)
except ToolError as e:
    if 'Unrecognized command' in str(e):
        reply_to_model_with_allowed_commands()
    else:
        raise

When it happens

Trigger: An LLM/tool caller passes command='delete_plan', 'createPlan', 'Add', ' remove' (whitespace), or any casing/typo variant; invoking the tool programmatically with a string not in the Literal set. Only exact lowercase matches dispatch.

Common situations: Agent frameworks where the model invents command names not in the tool schema; model output with casing or trailing whitespace; schema drift after upgrading the tool (a command removed or renamed) while prompts still reference the old name.

Related errors

• No primary agent available
• Parameter `plan_id` is required for command: create
• A plan with ID '{plan_id}' already exists. Use 'update' to m
• Parameter `title` is required for command: create
• Parameter `steps` must be a non-empty list of strings for co

AI-assisted analysis of FoundationAgents/OpenManus@52a13f2a57 (2026-08-15). Data as JSON: /api/errors/f589e1b393f76bde. Report an issue: GitHub.