aidputils.agents.toolkit.agent_name

aidputils.agents.toolkit.agent_name.add_inline_agent_name(message: langchain_core.messages.BaseMessage) → langchain_core.messages.BaseMessage

Add name and content XML tags to the message content.

Examples

>>> add_inline_agent_name(AIMessage(content="Hello", name="assistant"))
AIMessage(content="<name>assistant</name><content>Hello</content>", name="assistant")

>>> add_inline_agent_name(AIMessage(content=[{"type": "text", "text": "Hello"}], name="assistant"))
AIMessage(content=[{"type": "text", "text": "<name>assistant</name><content>Hello</content>"}], name="assistant")

aidputils.agents.toolkit.agent_name.remove_inline_agent_name(message: langchain_core.messages.BaseMessage) → langchain_core.messages.BaseMessage

Remove explicit name and content XML tags from the AI message content.

Examples

>>> remove_inline_agent_name(AIMessage(content="<name>assistant</name><content>Hello</content>", name="assistant"))
AIMessage(content="Hello", name="assistant")

>>> remove_inline_agent_name(AIMessage(content=[{"type": "text", "text": "<name>assistant</name><content>Hello</content>"}], name="assistant"))
AIMessage(content=[{"type": "text", "text": "Hello"}], name="assistant")

aidputils.agents.toolkit.agent_name.with_agent_name(model: langchain_core.language_models.LanguageModelLike, agent_name_mode: Literal['inline']) → langchain_core.language_models.LanguageModelLike

Attach formatted agent names to the messages passed to and from a language model.

This is useful for making a message history with multiple agents more coherent.

NOTE: agent name is consumed from the message.name field.

If you’re using an agent built with create_react_agent, name is automatically set. If you’re building a custom agent, make sure to set the name on the AI message returned by the LLM.

Parameters:

• model – Language model to add agent name formatting to.

• agent_name_mode –

  Use to specify how to expose the agent name to the LLM. - “inline”: Add the agent name directly into the content field of the AI message using XML-style tags.

  Example: “How can I help you” -> “<name>agent_name</name><content>How can I help you?</content>”.