openchronicle-mcp MCP Server
io.github.CSOAI-ORG/openchronicle-mcp
Persistent semantic + keyword memory database for LLM agents with project namespacing and hybrid search.
What is the openchronicle-mcp MCP server?
OpenChronicle is a memory database for LLM agents that persists decisions, milestones, and rejected approaches across sessions. It provides hybrid full-text and semantic search via Reciprocal Rank Fusion, project-scoped namespacing, git-onboarding for code context, and runs as a single ASGI process serving both HTTP REST and MCP transports.
OpenChronicle lets LLM agents maintain long-term memory that survives context compression and new conversations. It stores and retrieves information using both semantic and keyword search, organizes memory by project to prevent context leakage, can ingest git repository history to seed memory with code rationale, and degrades gracefully when embedding providers are unavailable. Run it on your own hardware as a single container or process.
How to install openchronicle-mcp
Copy-paste configuration for popular MCP clients.
Tools & capabilities
Tools this server exposes to the agent.
memory_save— Save a memory entry with optional tags and metadata to a project.memory_search— Search memory using hybrid full-text and semantic search via Reciprocal Rank Fusion.health— Check the health status of the memory database and embedding provider.
Use cases
- Persist architectural decisions and rejected approaches across multiple Claude conversations or coding sessions.
- Search project-specific context by semantic meaning or keyword to ground agent decisions in prior work.
- Onboard an LLM agent with git repository history by clustering commits and generating summaries for memory ingestion.
- Maintain separate memory namespaces for different projects or workstreams without context leakage.
- Degrade search to full-text only when embedding providers are unavailable, then backfill when service returns.
openchronicle-mcp MCP server FAQ
OpenChronicle is a persistent memory database designed for LLM agents. It stores decisions, milestones, and rejected approaches using both semantic and keyword search, organizes memory by project, and survives context compression across sessions.
Yes. OpenChronicle is licensed under AGPL-3.0 and is free and open-source software.
Install via pip (`pip install openchronicle-mcp`), run `oc init` and `oc serve` to start the server on localhost:8000, then register it with Claude Code using `claude mcp add --scope user --transport http openchronicle http://127.0.0.1:8000/mcp`.
Bearer-token authentication via `OC_API_KEY` is optional and disabled by default for trusted-LAN deployments. Enable it for untrusted networks; see the security posture documentation.
Yes. A Docker image is available at `ghcr.io/carldog/openchronicle-mcp`. Set `OC_API_HOST=0.0.0.0` in the container and mount volumes for data and config persistence.
Search degrades cleanly to full-text search only. The degraded state is reported via the health endpoint, and backfill catches up automatically when the provider returns.
README (reference)
Source of truth, from the repository.
OpenChronicle
<!-- markdownlint-disable MD033 --> <!-- fleet-confidence --> <sub>· claude-fable-5 · 2026-08-30 · details</sub>
A memory database for LLM agents. Persistent semantic + keyword memory, project namespacing, git-onboard, served over HTTP REST and MCP from a single ASGI process. Runs on your hardware.
What it does
- Persistent memory across sessions. Save decisions, milestones, and rejected approaches that survive context compression and new conversations. Retrieve them with hybrid full-text and semantic search via Reciprocal Rank Fusion.
- Project namespacing. Memory is scoped to projects, so context for one workstream doesn't leak into another.
- Git onboarding. Clone a repo, cluster commits by relatedness, return summaries ready for memory ingestion. Seeds long-term memory with the WHY behind existing code.
- One process, two transports. FastAPI hosts both the REST surface
(
/api/v1/*) and the MCP streamable-HTTP transport (/mcp) on the same port. Single container, single port mapping, single healthcheck. - Embedding-failure degradation. When the embedding provider goes
down, search degrades cleanly to FTS5-only and surfaces the
degraded state via
/api/v1/healthand the MCPhealthtool. Backfill catches up when the provider returns; the static/healthendpoint remains a minimal liveness probe. - Optional operational metrics. Released images (since v3.4.0) include
the bounded Prometheus recorder and guarded
/metricsendpoint, off by default. Opt in withOC_METRICS_ENABLED=true; enabling it in production stays subject to the performance gates. See the metrics configuration and the optional local monitoring runbook. - Schema migration framework. Versioned
.sqlmigrations with savepoint atomicity. Re-runs are idempotent. Future schema changes drop in asNNN_<slug>.sqlfiles. - Verified online backups. Uses SQLite's online backup API; each
nightly snapshot is published with a verified manifest in
OC_BACKUP_DIR, and one that fails verification is quarantined. Backup-before-destructive policy: vacuum runs a backup first as part of the same job. Integrity-check failures trigger emergency backups. - Optional encrypted offsite copies. A nightly job encrypts the newest snapshots with age and copies them to any rclone remote, append-only. See cloud_backup.md.
What it isn't
- Not a conversation engine. v3 has no LLM. Use Claude Code, Goose, Open WebUI, etc. via the MCP server.
- Not multi-tenant. Single user. Bearer-token auth via
OC_API_KEYis supported but optional — disabled by default for trusted-LAN deployments. Seedocs/configuration/security_posture.mdfor the when-to-enable guidance. - Not a cloud sync layer. The DB lives on your hardware. Backups go to
a local backup directory and, optionally, encrypted to a cloud remote,
as backup only. Cross-device sync isn't built in
(
docs/design/0001-cloud-backup.md).
By design.
Install
From source:
pip install -e ".[mcp,openai]"
oc init
oc serve
The default oc serve binds 127.0.0.1:8000. Override with
--host/--port or OC_API_HOST/OC_API_PORT.
Docker (single container, NAS-friendly):
docker run --rm \
-p 8000:8000 \
-e OC_API_HOST=0.0.0.0 \
-v $(pwd)/data:/app/data \
-v $(pwd)/config:/app/config \
ghcr.io/carldog/openchronicle-mcp:latest
OC_API_HOST=0.0.0.0 is required in a container — the app default
binds container-loopback, which the port mapping can't reach. To call
the server by anything other than localhost (a NAS hostname, a LAN
IP), also set OC_MCP_ALLOWED_HOSTS=your-host:* or every request gets
a 421 (see
env_vars.md).
For a Portainer stack on a NAS, use the docker-compose.nas.yml at
the repo root. It needs three things first: OC_TAG set to a release tag
(there is no :latest fallback), the data volume created once
(docker volume create openchronicle-mcp_oc-data; the compose never
creates it, so a missing volume fails the deploy instead of starting
empty), and the host exports directory created and owned by uid 1000. The
file's header comments list every variable.
Quickstart
# Bootstrap the runtime tree
oc init
# Create a project
PROJECT_ID=$(oc init-project "my-project")
# Save your first memory
oc memory add "Decision: SQLite for storage; AGPL for license" \
--project-id $PROJECT_ID --tags decision
# Search it
oc memory search "storage decision" --project-id $PROJECT_ID
Or do the same via MCP — register the server with Claude Code:
claude mcp add --scope user --transport http openchronicle \
http://127.0.0.1:8000/mcp
Then ask Claude to call memory_save and memory_search.
Architecture
Hexagonal: domain/ (pure types + ports) → application/ (use cases,
services) → infrastructure/ (SQLite, embedding adapters, the
maintenance loop). Driver-side adapters in interfaces/ host the
HTTP, MCP, and CLI surfaces.
See docs/architecture/ARCHITECTURE.md for the full layout.
Documentation
docs/architecture/ARCHITECTURE.md— layout, schema, ASGI designdocs/architecture/MAINTENANCE.md— maintenance loop + degradation policydocs/cli/commands.md—ocsubcommand referencedocs/configuration/env_vars.md— environment variablesdocs/configuration/config_files.md—core.jsonschemadocs/configuration/security_posture.md— security modeldocs/integrations/mcp_client_setup.md— register the MCP serverdocs/integrations/mcp_server_spec.md— MCP tool surfacedocs/api/STABILITY.md— versioning + deprecation policydocs/design/README.md— proposed designs and comparative repository reviews
Development
pip install -e ".[dev,mcp,openai,ollama]"
pre-commit install
pytest
The architecture is enforced by tests:
tests/test_hexagonal_boundaries.py— domain/application/infrastructure layeringtests/test_architectural_posture.py— core agnostic of MCP SDKtests/test_no_secrets_committed.py,tests/test_no_soft_deprecation.py— repo hygiene
License
Copyright (C) 2025-2026 CarlDog
AGPL-3.0. This program is free software: you can redistribute it and/or modify it under the terms of the GNU Affero General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. It is distributed WITHOUT ANY WARRANTY; see the license for details.
The copyright line lives here rather than inside LICENSE: that file is
the AGPL text verbatim, and the <year> <name of author> placeholders in
its closing appendix are the license's own instructions for what to put
in your source files — not blanks to fill in. Editing them would modify
the license text itself.
Related MCP servers
Optometry Ai Safety MCP Server by MEOK AI Labs
FHIR-based patient records for domiciliary opticians
Generate machine-readable NIST OSCAL packages (SSP/component-definition) + FedRAMP RFC-0024
