Skip to main content
Owner guard on edits, Context Retrieval for users only

Keep an agent inside one user’s memories

  • Optional user_id on update and delete_by_ids
    Pass the signed-in user’s id and the call only touches that user’s memories. update returns 404 for another user’s memory, exactly like a missing one; delete_by_ids skips another user’s ids. Set it from your own session, never from model output.
  • delete_by_ids returns deleted
    The response now counts the memories the request actually removed.
  • Context Retrieval requires user_id
    POST /api/v2/memory/context searches one user’s memories. user_id is required, and agent_id is rejected with 422 instead of returning an empty result. Use search for agent memories.
Memory space purge

Start a memory space over

POST /api/v2/memory/purge permanently deletes all memory data in a memory space, or in one app or project inside it, so you can reuse it from empty. There is no undo.
  • Two confirmed steps
    The first call deletes nothing: it returns a nonce and a count of what would be deleted. Only POST /api/v2/memory/purge/confirm with that nonce, within 5 minutes, starts the purge.
  • Runs as a job
    Poll GET /api/v2/memory/purge/{job_id} for progress per store. While it runs, writes to the memory space return 409 purge_in_progress; reads keep working.
  • Scoped
    Purge the whole memory space, one app or one project. Files uploaded through /api/v2/object/sign are not deleted.
See Purge a Memory Space.
Context Retrieval

Retrieval that keeps searching until it has the answer

POST /api/v2/memory/context hands retrieval to an LLM. It calls keyword, semantic and hybrid search, reads whole conversations when it needs to, decides what is still missing, and searches again, until the evidence is sufficient. You get back the memories it found, not an answer.
  • Same request and response as search
    Send the body you already send to /search. The response is the same SearchData, so existing parsing code works unchanged.
  • See what the answer rests on
    Episodes the model relied on come back with cited: true, ranked first.
  • Bounded
    At most 5 rounds and 45 seconds per request. Reaching a limit returns what was found so far, never an error.
Use it for multi-hop, temporal and cross-conversation questions. See Context Retrieval.
Memory editing and feedback

Fix what EverOS remembers

Memories are no longer read-only. Three new endpoints let you correct, remove and rate individual memories.
  • Correct an episode
    POST /api/v2/memory/update rewrites an episode’s narrative, summary or subject. A changed narrative is re-indexed at once, and its facts and the profile are refreshed in the background. Your edit survives later consolidation.
  • Remove specific memories
    POST /api/v2/memory/delete_by_ids deletes up to 50 memories by id, together with the facts derived from them. Deleting a whole user, agent or session works as before.
  • Tell us when a memory is wrong
    POST /api/v2/memory/feedback records a positive or negative rating on an episode, profile item, agent case or agent skill, with an optional reason and a suggested correction. Recording changes nothing by itself.
  • See what was edited
    Episodes edited by hand now carry edited_at and reflect_state in get and search responses.
See the API Reference for the three endpoints.
v1.0.0

Unified v2 Memory API: one contract, Cloud or self-hosted

EverOS Cloud now runs on the same v2 Memory API as open-source EverOS. Prototype locally, ship on managed Cloud, or switch between them for compliance, without rewriting your integration or getting locked into a black box.
  • Write once, run either place
    add, flush, get, and search behave identically on Cloud and self-hosted.
  • Multi-party memory out of the box
    Every message carries a sender_id, so group chats and human-plus-agent conversations attribute memory to the right participant automatically.

Retrieve the right context

  • Four retrieval strategies
    Choose keyword, vector, hybrid, or agentic search per query, and scope any retrieval to one user_id or agent_id with time and field filters.

Agent memory, not just user memory

  • Cases and skills as first-class memory types
    Alongside user profiles and episodes, retrieve what an agent has learned about itself, so your agents get better at a task the more they run it.

Full lifecycle control

Memory is a governable asset, not a write-once black box.
  • Edit a user’s profile directly; your manual edits survive future background consolidation.
  • Delete by user, agent, or session to meet privacy and retention needs.

Compatibility

  • The v1 API remains available for existing integrations.
See the Cloud Quickstart and API Reference.
v0.4.0

Request Log Improvements:

  • Find requests faster with request ID filtering, custom time-range search, and pagination.
  • Click any entry to view its status, latency, and error messages.

API v0 Retirement:

  • The v0 API is now fully replaced by v1. Calls to v0 endpoints will return an error.
  • Existing data is migrated automatically, so users already on v1 are unaffected.
  • See the “API Reference (Cloud)” for the migration guide.
v0.3.1

Streamlined Onboarding:

  • Simpler, friendlier setup
    A more streamlined onboarding experience that’s friendlier for users of any technical level.
  • Helpful tips while you wait
    While your memory space is being set up, we share helpful tips tailored to you to get you started.
  • Refreshed look
    A cleaner, fully branded onboarding experience.
v0.3.0

Console UI/UX Refresh:

  • Zero-setup onboarding
    Copy a single prompt, paste it into any agent, and you’re done. No manual install, no API key creation, all automated.
  • Redesigned Get Started page
    Switch between Prompt (for agents) and SDK setup modes, with ready-to-use add / get / search code samples for both Agent and User memory.
  • Clearer navigation
    Reorganized sidebar tabs so you can find what you need faster.
v0.2.1

Performance Improvmenets:

  • System-level engineering optimizations for improved speed and stability
v0.2.0

New Features & Updates:

  • API path set to v1
    The EverOS API has been upgraded to v1 with improved structure and new capabilities. See the “API Reference” for full details and migration guide
  • Multimodal storage and retrieval (mRAG)
    Store and query images, documents, PDFs and web contents alongside text
  • Self-evolving agent memory
    Agents can now build and refine memory through Cases and Skills primitives
  • Memory Bank
    Visualize and manage user, group, and agent memory from a single console
  • Interactive Playground
    Explore and test memory APIs directly in the console, no setup required
v0.1.3

New Features & Updates:

  • Dashboard UI updates
  • Performance improvements
  • Bug fixes and reliability improvements
v0.1.2

New Features & Updates:

  • Message timestamp validation

Bug Fixes:

  • Fixed dashboard API usage chart data anomaly
v0.1.1

New Features & Updates:

  • Refactored Get Memories and Search Memories API parameters
  • Dashboard UI interaction improvements
  • Added Google and GitHub third-party login support
  • Restricted available LLM providers and models on cloud platform
  • Bug fixes
This is a non-backwards-compatible change. Update your API request parameters immediately.
v0.1.0

New Features & Updates:

  • API path set to v0
    All endpoints are served under /api/v0/. Requests to any other path are not supported.