Independent system
ocbrain
The thing I want from agent memory is not a longer prompt. I want a system that can say what it believes, show where the belief came from, notice when the result was harmful, and stop serving the belief when the evidence changes.
ocbrain v1.1.0 is that system in a deliberately small shape: one local SQLite ledger shared by Codex, Claude Code, OpenClaw, and compatible clients through MCP.
The basic shape
One brain, several lenses.
Each runtime already produces useful history, but separate memories compound in separate directions. ocbrain brings that evidence into one ledger. Project, repo, task, client, runtime, and session context decide which part of the ledger a given agent may see.
That gives the system a shared history without pretending all history belongs everywhere. A stable operating principle can travel. A confidential client fact cannot.
The decision loop
Evidence becomes knowledge only through outcomes.
Evidence
Immutable, hash-pinned records preserve what happened and where it came from.
Knowledge
Current beliefs compile from evidence and keep support, contradiction, scope, and lifecycle attached.
Feedback
Corrections, tests, task outcomes, and retrieval usefulness decide what earned trust.
Memory
A bounded view serves only current injectable knowledge that survived the applicable safeguards.
The autonomy belongs to maintaining this loop. ocbrain does not run the user's work, install skills, publish packages, or send a hosted request on its own.
Agents finish substantive work with retrieval feedback and a verified closeout. The receipt connects the evidence they saw to the action they took and the outcome that followed.
Across the current runtimes
Connection is not the same as proof.
The current install has been exercised from Codex, Claude Code, and OpenClaw against the same v1 core. A real integration means a fresh client can retrieve bounded context, expand a hash-verified source, use it, and attach the eventual result to the same evidence trail.
The canonical install page uses the runtimes' current MCP commands and ends with a model-driven tool-use check. That is the level where an integration becomes real.
The long game
The memory corpus should eventually train the model.
The same history can compile into local evaluation and training streams, but the evidence ledger remains authoritative. Derived embeddings, summaries, rankings, and models are replaceable; source identity, scope, corrections, actions, and outcomes are not.
Retrieval quality is not the finish line. A result matters only when the context was relevant, changed the decision, and survived contact with the eventual outcome.
That is why v1.1.0 records retrieval usefulness and verified closeouts as part of the product loop. Better priors should be earned from what happened next rather than inferred from what looked plausible in the prompt.
Why this shape
Models rotate. A legible decision history compounds.
My instinct is to keep the brain simple enough to inspect: code enforces the trust boundary, SQLite keeps the graph and evidence, and models help where semantic judgment is actually useful. Once that loop is right, the retrieval, grading, and training policy can improve without changing what the system owes the user.
It is not magical, but it is trustworthy. And trust is what lets the system compound.