What causes "Agent is already bound to a different ID" in microsoft/autogen?

Create a fresh agent instance per binding instead of rebinding an existing one.

microsoft/autogen · error · RuntimeError

Agent is already bound to a different ID

Error message

Agent is already bound to a different ID

What it means

bind_id_and_runtime(id, runtime) assigns an identity to an agent. If the agent already has an _id (set either in __init__ during runtime factory instantiation or a prior bind) and the new id differs (compared with AgentId equality: same type and key), it raises RuntimeError. This prevents rebinding one agent instance to a different identity.

Source

Thrown at python/packages/autogen-core/src/autogen_core/_base_agent.py:96

        return cls.internal_unbound_subscriptions_list

    @property
    def metadata(self) -> AgentMetadata:
        assert self._id is not None
        return AgentMetadata(key=self._id.key, type=self._id.type, description=self._description)

    def __init__(self, description: str) -> None:
        if AgentInstantiationContext.is_in_factory_call():
            self._runtime: AgentRuntime = AgentInstantiationContext.current_runtime()
            self._id = AgentInstantiationContext.current_agent_id()
        if not isinstance(description, str):
            raise ValueError("Agent description must be a string")
        self._description = description

    async def bind_id_and_runtime(self, id: AgentId, runtime: AgentRuntime) -> None:
        if hasattr(self, "_id"):
            if self._id != id:
                raise RuntimeError("Agent is already bound to a different ID")

        if hasattr(self, "_runtime"):
            if self._runtime != runtime:
                raise RuntimeError("Agent is already bound to a different runtime")

        self._id = id
        self._runtime = runtime

    @property
    def type(self) -> str:
        return self.id.type

    @property
    def id(self) -> AgentId:
        return self._id

    @property
    def runtime(self) -> AgentRuntime:

View on GitHub (pinned to 027ecf0a37)

Solutions

Create a fresh agent instance per binding instead of rebinding an existing one.
If the id is already correct (set during factory instantiation), skip calling bind_id_and_runtime.
Check agent.id first and only bind when unbound or identical.

Example fix

# before
agent = MyAgent("d")
await agent.bind_id_and_runtime(AgentId("t", "k1"), runtime)
await agent.bind_id_and_runtime(AgentId("t", "k2"), runtime)  # RuntimeError

# after
agent2 = MyAgent("d")
await agent2.bind_id_and_runtime(AgentId("t", "k2"), runtime)

Defensive patterns

Strategy: validation

Validate before calling

if getattr(agent, "_id", None) is not None and agent._id != new_id:
    raise RuntimeError(f"agent already bound to {agent._id}")
await agent.bind_id_and_runtime(new_id, runtime)

Prevention

Bind each agent exactly once; create new instances for new IDs.
Prefer factory registration so the runtime assigns IDs and you never call bind_id_and_runtime manually.
Track construction path in code review: factory-instantiated agents already have _id set.

When it happens

Trigger: Calling await agent.bind_id_and_runtime(AgentId('a','1'), runtime) on an agent that was already instantiated by a runtime with AgentId('a','2'); manually binding the same agent object twice with different IDs; re-registering an existing agent instance under a new type/key.

Common situations: Caching agent instances and reusing them across registrations; test code that binds one agent to multiple runtimes/ids; subclass code that calls bind after the factory already set the id.

Related errors

AI-assisted analysis of microsoft/autogen@027ecf0a37 (2026-08-15). Data as JSON: /api/errors/fde69e1af2648410. Report an issue: GitHub.