Search Archival Memory

Deprecated

agents.passages.search(, ) -> PassageSearchResponse

get/v1/agents/{agent_id}/archival-memory/search

Search archival memory using semantic (embedding-based) search with optional temporal filtering.

This endpoint allows manual triggering of archival memory searches, enabling users to query an agent's archival memory store directly via the API. The search uses the same functionality as the agent's archival_memory_search tool but is accessible for external API usage.

ParametersExpand Collapse

agent_id: str

The ID of the agent in the format 'agent-'

minLength42

maxLength42

query: str

String to search for using semantic similarity

end_datetime: Optional[Union[str, datetime, null]]

Filter results to passages created before this datetime

formatdate-time

start_datetime: Optional[Union[str, datetime, null]]

Filter results to passages created after this datetime

formatdate-time

tag_match_mode: Optional[Literal["any", "all"]]

How to match tags - 'any' to match passages with any of the tags, 'all' to match only passages with all tags

Accepts one of the following:

"any"

"all"

tags: Optional[SequenceNotStr[str]]

Optional list of tags to filter search results

top_k: Optional[int]

Maximum number of results to return. Uses system default if not specified

ReturnsExpand Collapse

class PassageSearchResponse: …

count: int

Total number of results returned

results: List[Result]

List of search results matching the query

content: str

Text content of the archival memory passage

timestamp: str

Timestamp of when the memory was created, formatted in agent's timezone

tags: Optional[List[str]]

List of tags associated with this memory

Search Archival Memory

from letta_client import Letta

client = Letta(
    api_key="My API Key",
)
response = client.agents.passages.search(
    agent_id="agent-123e4567-e89b-42d3-8456-426614174000",
    query="query",
)
print(response.count)

{
  "count": 0,
  "results": [
    {
      "content": "content",
      "timestamp": "timestamp",
      "tags": [
        "string"
      ]
    }
  ]
}

Returns Examples

{
  "count": 0,
  "results": [
    {
      "content": "content",
      "timestamp": "timestamp",
      "tags": [
        "string"
      ]
    }
  ]
}