EVevermind-ai
memory

Post Search

Changed on

Hybrid retrieval across the configured memory backends.

post/api/v1/memory/search

Request

  • The document declares no server URL.
  • Auth
    none declared

Request body

user_idstring nullable
agent_idstring nullable
app_idstring
project_idstring
querystring required
method'keyword' | 'vector' | 'hybrid' | 'agentic' | 'llm_multiround'

Public method enum. RRF / LR / vector_anchored are hidden under HYBRID.

top_kinteger
radiusnumber nullable
min_scorenumber nullable
include_profileboolean
enable_llm_rerankboolean

Opt-in LLM rerank pass for HYBRID. Applies to agent_case and agent_skill fusion only; the episode hybrid path has built-in fact eviction and ignores this flag. Ignored by keyword / vector / agentic.

Response

Successful Response

request_idstring required

Changes

  • bf705e3f32e2Revision changes
    • info

      added the new llm_multiround enum value to the request property

  • 65f2f4743b36Revision changes
    • breaking

      the response property /// became optional for the status

    • breaking

      response property data/episodes/items/session_id list-of-types was widened by adding types null to media type application/json of response 200

    • info

      added the new optional request property