memory

Load Pinned

Deterministically load a context's always-delivery memories (Issue #886).

The deterministic counterpart to /recall: returns the complete, unranked, ordered delivery_mode='always' set for the given context — no semantic ranking, no rerank. Items carry L1 + L2 only (summary + context_summary); fetch full content via /reference. Bounded by settings.pinned_load_cap (or the request cap); when the pinned set exceeds the cap, truncated is true and total_available reports the real count — never a silent truncation.

Request: {"context_id": "<context-uuid>", "cap": 100}

Example: POST /api/v1/memory/pinned Authorization: Bearer <session_token>

post/api/v1/memory/pinned

Request

  • Base URL: https://memory.kagura-ai.com (the document declares no server; this is the origin it was published from)
  • URL: https://memory.kagura-ai.com/api/v1/memory/pinned
  • Auth: none declared

Headers

authorizationstring nullable

Request body

context_idstring nullable

Context UUID whose pinned memories to load

capinteger nullable

Optional override for the max returned (bounded)

Response

Successful Response

statusstring
total_availableinteger required
truncatedboolean required
capinteger required

Changes

No recorded changes to this operation across all 1 revision of this API.