Local-first, source-grounded memory that survives AI agent sessions.
MCP Server: io.github.ampres-ai/talamus
The MCP server io.github.ampres-ai/talamus provides local-first, source-grounded memory designed to persist across AI agent sessions. Its description indicates it is focused on maintaining memory state on the local environment while staying grounded in source material.
🛠️ Key Features
Local-first memory
Source-grounded memory
Persists across AI agent sessions
🚀 Use Cases
Retaining information between multiple runs of AI agent sessions
Keeping agent memory aligned with source material over time
⚡ Developer Benefits
Clear persistence model for session-to-session continuity
Emphasis on source-grounded memory for more traceable context
⚠️ Limitations
No additional details provided about tools, interfaces, or configuration in the available server data
Your coding agent forgets why a decision was made as soon as the session ends.
Talamus keeps the decisions, evidence, and corrections worth remembering as
ordinary Markdown, then gives Claude Code, Codex, Cursor, Gemini CLI, and any
MCP agent cited recall in the next session.
No hosted account. No telemetry. No required embeddings. Plain search stays
on your machine; LLM-backed actions use only the engine you choose.
Try the whole local retrieval loop first — no persistent install, account,
LLM, or hook, and no files written outside ./talamus-demo:
Set up the project brain. talamus setup initializes the brain, chooses an engine, installs MCP for Claude Code, Cursor, Codex, OpenCode, and OpenClaw when detected, asks once before installing the session-capture hook, and can probe the engine with one tiny live call.
bash
talamus setup
Your agent session ends. The consented hook reads the transcript and git diff, applies the worth-remembering gate, writes only useful memory into this brain, and audits the event at .talamus/logs/capture.log.
A fresh session asks what happened and gets an answer from real notes, with sources.
bash
talamus recall "why did we choose FTS5?"
talamus ask "why did we choose FTS5?"
Reproduce the scripted demo without spending LLM calls, or run it with your real engine.
Also measured in committed artifacts: −97.7% tokens per answer versus loading the brain into context, refusal 1.000 on out-of-scope questions, and search latency p95 72.6 ms at 10k notes / p50 624 ms at 100k.
The honest part: retrieval quality tracks the LLM you bring. With a strong expansion engine, talamus-smart leads a strong multilingual dense model (multilingual-e5) on every metric including ranking (nDCG 0.847 vs 0.837); with a weak or free one, e5 leads ranking while Talamus keeps the best hit/recall — and on a slow local engine, plain search beats --smart outright. Every number traces to a committed artifact; the losses stay on the table.
Engines
Bring the LLM you already have: claude-cli, codex-cli, antigravity-cli (agy), opencode, ollama, or anthropic-api.
Quickstart
bash
pipx install "talamus[mcp]"
talamus setup
talamus ingest ./notes && talamus ask "what should I remember?"
Run talamus for the status dashboard, talamus quickstart for essential commands, or talamus ui for the local React workbench.
Install the consent-aware Talamus agent skill from skills.sh:
OpenClaw can install the same standalone skill directly from ClawHub:
bash
openclaw skills install @ampres-ai/talamus-memory
Installing the standalone skill does not install Talamus automatically. If the
CLI is missing, the skill explains the isolated installation choices and asks
before running one.
Gemini CLI can install Talamus directly from its extension gallery or from this
repository. The extension starts the pinned PyPI release through uvx, so it
does not modify the cloned source tree:
goose can install the repository as an Open Plugin. This adds the consent-aware
memory skill and starts the pinned local MCP server for each new CLI session:
