langgraph_agent_toolkit.core.observability.empty module

class langgraph_agent_toolkit.core.observability.empty.EmptyObservability(remote_first=False)[source][source]

Bases: BaseObservabilityPlatform

Observability platform with in-memory prompt storage.

This supports prompt templates without a remote Langfuse or LangSmith backend.

Initialize EmptyObservability.

Parameters:

remote_first (bool) – Ignored by this implementation.

__init__(remote_first=False)[source][source]

Initialize EmptyObservability.

Parameters:

remote_first (bool) – Ignored by this implementation.

get_callback_handler(**kwargs)[source][source]

Get the callback handler for the observability platform.

Return type:

None

before_shutdown()[source][source]

Perform any necessary cleanup before shutdown.

Return type:

None

record_feedback(run_id, key, score, **kwargs)[source][source]

Record feedback by ignoring it without a remote backend.

Parameters:

  • run_id (str)

  • key (str)

  • score (float)

Return type:

None

push_prompt(name, prompt_template, metadata=None, force_create_new_version=True)[source][source]

Store a prompt in memory.

Parameters:

  • name (str) – Prompt name.

  • prompt_template (str | List[ChatMessageDict]) – Prompt template to store.

  • metadata (Dict[str, Any] | None) – Optional prompt metadata.

  • force_create_new_version (bool) – Overwrite an existing prompt when True.

Return type:

None

pull_prompt(name, template_format='f-string', **kwargs)[source][source]

Retrieve a prompt from memory.

Parameters:

  • name (str) – Prompt name.

  • template_format (Literal['f-string', 'mustache', 'jinja2']) – Template format for processing.

  • **kwargs – Ignored arguments.

Returns:

Stored prompt template as a ChatPromptTemplate.

Raises:

ValueError – If the prompt is missing from memory.

Return type:

ChatPromptTemplate | str | dict | None

delete_prompt(name)[source][source]

Delete a prompt from memory.

Parameters:

name (str) – Prompt name to delete.

Return type:

None