OpenBMB/ChatDev · error · ValueError

Tool name '{spec.name}' conflicts with a built-in skill tool

Error message

Tool name '{spec.name}' conflicts with a built-in skill tool

What it means

Raised when merging skill tool specs into the agent's tool list: a skill-provided tool has the same name as one of the built-in/baseline tools already present. The executor refuses duplicate tool names because the model would have ambiguous tool targets.

Source

Thrown at runtime/node/executor/agent_executor.py:304

                )

        if not parts:
            return None
        return "\n\n".join(part for part in parts if part)

    def _merge_skill_tool_specs(
        self,
        tool_specs: List[ToolSpec],
        skill_manager: AgentSkillManager | None,
    ) -> List[ToolSpec]:
        if skill_manager is None:
            return tool_specs

        merged = list(tool_specs)
        existing_names = {spec.name for spec in merged}
        for spec in skill_manager.build_tool_specs():
            if spec.name in existing_names:
                raise ValueError(f"Tool name '{spec.name}' conflicts with a built-in skill tool")
            existing_names.add(spec.name)
            merged.append(spec)
        return merged

    def _build_agent_invoker(
        self,
        provider: ModelProvider,
        client: Any,
        base_call_options: Dict[str, Any],
        default_tool_specs: List[ToolSpec],
        node: Node,
    ) -> Callable[[List[Message]], Message]:
        """Create a callable that other components can use to invoke the model."""

        def invoke(
            conversation: List[Message],
            *,
            tools: Optional[List[ToolSpec]] = None,

View on GitHub (pinned to 4fb2db0ea9)

Solutions

1. Rename your custom tool so it does not collide with names returned by AgentSkillManager.build_tool_specs() (e.g. 'activate_skill', 'read_skill_file')
2. Filter your tool_specs against skill_manager.build_tool_specs() names before passing them in
3. Disable Agent Skills for the node if you intend to provide your own tools with those names

Example fix

# before
tools = [{"name": "activate_skill", ...}]  # collides

# after
tools = [{"name": "my_activate_skill", ...}]

skill_names = {s.name for s in skill_manager.build_tool_specs()}
assert not ({t['name'] for t in tool_specs} & skill_names), 'name collision'

When it happens

Trigger: Calling the agent executor with tool_specs containing a tool named 'activate_skill' or 'read_skill_file' (or any name AgentSkillManager.build_tool_specs() also returns) while Agent Skills are enabled for the node.

Common situations: Custom tool list built by a user that happens to include a tool named like a skill tool; upgrading to a runtime version that newly enables Agent Skills; copy-pasting tool names from skill docs into your own tool registry.

Related errors

• Agent Skills are not enabled for this node
• No node types registered; cannot validate workflow
• '{legacy_key}' is deprecated. Use the new graph-level memory
• missing frontmatter
• missing closing delimiter

AI-assisted analysis of OpenBMB/ChatDev@4fb2db0ea9 (2026-08-27). Data as JSON: /api/errors/10d1b4933aac5479. Report an issue: GitHub.