Skillget

ogham-mcp

mcp
SUMMARY

Shared memory MCP server — persistent, searchable, cross-client

README.md

Ogham MCP

Ogham (pronounced "OH-um") -- persistent, searchable shared memory for AI coding agents. Works across clients.

License: MIT
Docker
Python 3.13+
PyPI

Contents

Retrieval quality

97.2% Recall@10 on LongMemEval (500 questions, ICLR 2025). No LLM in the search pipeline -- one PostgreSQL query with Reciprocal Rank Fusion.

0.69 R@10 on BEAM (400 questions across 10 memory abilities, ICLR 2026). With optional FlashRank reranking: 0.70 R@10. Full benchmark.

End-to-end QA accuracy on LongMemEval (retrieval + LLM reads and answers):

System Accuracy Architecture
OMEGA 95.4% Classification + extraction pipeline
Observational Memory (Mastra) 94.9% Observation extraction + GPT-5-mini
Hindsight (Vectorize) 91.4% 4 memory types + Gemini-3
Zep (Graphiti) 71.2% Temporal knowledge graph + GPT-4o
Mem0 49.0% RAG-based

Retrieval only (R@10 -- no LLM in the search loop):

System R@10 Architecture
Ogham 97.2% 1 SQL query (pgvector + tsvector CCF hybrid search)
LongMemEval paper baseline 78.4% Session decomposition + fact-augmented keys

Other retrieval systems that report similar R@10 numbers typically use cross-encoder reranking, NLI verification, knowledge graph enrichment, and LLM-as-a-judge pipelines. Ogham reaches 97.2% with one Postgres query. Optional FlashRank reranking is available for self-hosters who want extra ranking precision.

These tables measure different things. QA accuracy tests whether the full system (retrieval + LLM) produces the correct answer. R@10 tests whether retrieval alone finds the right memories. Ogham is a retrieval engine -- it finds the memories, your LLM reads them.

Category R@10 Questions
single-session-assistant 100% 56
knowledge-update 100% 78
single-session-user 98.6% 70
multi-session 97.3% 133
single-session-preference 96.7% 30
temporal-reasoning 93.5% 133

Full breakdown: ogham-mcp.dev/features

The problem

AI coding agents forget everything between sessions. Switch from Claude Code to Cursor to Kiro to OpenCode and context is lost. Decisions, gotchas, architectural patterns -- gone. You end up repeating yourself, re-explaining your codebase, re-debugging the same issues.

Ogham gives your agents a shared memory that persists across sessions and clients.

Quick start

1. Install

uvx ogham-mcp init

This runs the setup wizard. It walks you through everything: database connection, embedding provider, schema migration, and writes MCP client configs for Claude Code, Cursor, VS Code, and others.

You need a database before running this. Either create a free Supabase project or a Neon database. The wizard handles the rest.

Using Neon or self-hosted Postgres? Install with the postgres extra so the driver is available:

uvx --from 'ogham-mcp[postgres]' ogham-mcp init

2. Add to your MCP client

The wizard configures everything and writes your client config -- including all environment variables the server needs. For Claude Code, it runs claude mcp add automatically. For other clients, copy the config snippet it prints.

3. Use it

Tell your agent to remember something, then ask about it later -- from the same client or a different one. It works because they all share the same database.

Manual setup (if you prefer)

If you'd rather configure things yourself instead of using the wizard:

# Supabase
export SUPABASE_URL=https://your-project.supabase.co
export SUPABASE_KEY=your-service-role-key
export EMBEDDING_PROVIDER=openai  # or ollama, mistral, voyage
export OPENAI_API_KEY=sk-...      # for your chosen provider

# Or Postgres (Neon, self-hosted)
export DATABASE_BACKEND=postgres
export DATABASE_URL=postgresql://user:pass@host/db
export EMBEDDING_PROVIDER=openai
export OPENAI_API_KEY=sk-...

Run the schema migration (sql/schema.sql for Supabase, sql/schema_postgres.sql for Neon/self-hosted), then add the MCP server to your client.

Installation methods

Method Command When to use
uvx (recommended) uvx ogham-mcp Quick setup, auto-updates
Docker docker pull ghcr.io/ogham-mcp/ogham-mcp Isolation, self-hosted
Git clone git clone + uv sync Development, contributions

Claude Code

claude mcp add ogham -- uvx ogham-mcp

OpenCode

Add to ~/.config/opencode/opencode.json:

{
  "mcp": {
    "ogham": {
      "type": "local",
      "command": ["uvx", "ogham-mcp"],
      "environment": {
        "SUPABASE_URL": "https://your-project.supabase.co",
        "SUPABASE_KEY": "{env:SUPABASE_KEY}",
        "EMBEDDING_PROVIDER": "openai",
        "OPENAI_API_KEY": "{env:OPENAI_API_KEY}"
      }
    }
  }
}

Docker

docker run --rm \
  -e SUPABASE_URL=https://your-project.supabase.co \
  -e SUPABASE_KEY=your-key \
  -e EMBEDDING_PROVIDER=openai \
  -e OPENAI_API_KEY=sk-... \
  ghcr.io/ogham-mcp/ogham-mcp

From source

git clone https://github.com/ogham-mcp/ogham-mcp.git
cd ogham-mcp
uv sync
uv run ogham --help

SSE transport (multi-agent)

By default, Ogham runs in stdio mode -- each MCP client spawns its own server process. For multiple agents sharing one server, use SSE mode:

ogham serve --transport sse --port 8742

The server runs as a persistent background process. All clients connect to the same instance -- one database pool, one embedding cache, shared memory.

Client config for SSE (any MCP client):

{
  "mcpServers": {
    "ogham": {
      "url": "http://127.0.0.1:8742/sse"
    }
  }
}

Health check at http://127.0.0.1:8742/health (cached, sub-10ms).

Configure via env vars (OGHAM_TRANSPORT=sse, OGHAM_HOST, OGHAM_PORT) or CLI flags. The init wizard (ogham init) walks through SSE setup if you choose it.

Entry points

Ogham has two entry points:

  • ogham -- the CLI. Use this for ogham init, ogham health, ogham search, and other commands you run yourself. Running ogham with no arguments starts the MCP server.
  • ogham-serve -- starts the MCP server directly. This is what MCP clients should call. When you run uvx ogham-mcp, it invokes ogham-serve.

CLI

ogham init                      # Interactive setup wizard
ogham health                    # Check database + embedding provider
ogham search "query"            # Search memories from the terminal
ogham store "some fact"         # Store a memory
ogham list                      # List recent memories
ogham profiles                  # List profiles and counts
ogham stats                     # Profile statistics
ogham export -o backup.json     # Export memories
ogham import backup.json        # Import memories
ogham cleanup                   # Remove expired memories
ogham hooks install             # Auto-detect client + configure hooks
ogham hooks recall              # Read from the stone (load project context)
ogham hooks inscribe            # Carve into the stone (capture activity)
ogham serve                     # Start MCP server (stdio, default)
ogham serve --transport sse     # Start SSE server on port 8742
ogham openapi                   # Generate OpenAPI spec

Configuration

Variable Required Default Description
DATABASE_BACKEND No supabase supabase or postgres
SUPABASE_URL If supabase -- Your Supabase project URL
SUPABASE_KEY If supabase -- Supabase secret key (service_role)
DATABASE_URL If postgres -- PostgreSQL connection string
EMBEDDING_PROVIDER No ollama ollama, openai, mistral, voyage, gemini, or onnx
EMBEDDING_DIM No 512 Vector dimensions -- must match your schema (see below)
OPENAI_API_KEY If openai -- OpenAI API key
MISTRAL_API_KEY If mistral -- Mistral API key
VOYAGE_API_KEY If voyage -- Voyage AI API key
GEMINI_API_KEY If gemini -- Google Gemini API key
OLLAMA_URL No http://localhost:11434 Ollama server URL
OLLAMA_EMBED_MODEL No embeddinggemma Ollama embedding model
MISTRAL_EMBED_MODEL No mistral-embed Mistral embedding model
VOYAGE_EMBED_MODEL No voyage-4-lite Voyage embedding model
GEMINI_EMBED_MODEL No gemini-embedding-2-preview Gemini embedding model
RERANK_ENABLED No false Enable FlashRank cross-encoder reranking
RERANK_ALPHA No 0.55 Cross-encoder score weight (0-1)
DEFAULT_MATCH_THRESHOLD No 0.7 Similarity threshold (see below)
DEFAULT_MATCH_COUNT No 10 Max results per search
DEFAULT_PROFILE No default Memory profile name

Embedding providers

Provider Default dimensions Recommended threshold Notes
OpenAI 512 (schema default) 0.35 Set EMBEDDING_DIM=512 explicitly -- OpenAI defaults to 1024
Ollama 512 0.70 Tight clustering, scores run 0.8-0.9
Mistral 1024 0.60 Fixed 1024 dims, can't truncate. Schema must be vector(1024)
Voyage 512 (schema default) 0.45 Moderate spread
Gemini 512 0.35 gemini-embedding-2-preview, supports MRL truncation
ONNX 1024 0.35 Local BGE-M3 inference, dense + sparse vectors. See ONNX section

EMBEDDING_DIM must match the vector(N) column in your database schema. The default schema uses vector(512). If you use Mistral, you need to alter the column to vector(1024) before storing anything.

Each provider clusters vectors differently, so the similarity threshold matters. Start with the recommended value and adjust based on your results.

Temporal search

Search queries with time expressions like "last week" or "three months ago" are resolved automatically using parsedatetime -- no configuration needed. This handles roughly 80% of temporal queries at zero cost.

For expressions that parsedatetime cannot parse ("the quarter before last", "around Thanksgiving"), set TEMPORAL_LLM_MODEL to call an LLM as a fallback:

# Self-hosted with Ollama (free, local)
TEMPORAL_LLM_MODEL=ollama/llama3.2

# Cloud API
TEMPORAL_LLM_MODEL=gpt-4o-mini

Any litellm-compatible model string works -- deepseek/deepseek-chat, moonshot/moonshot-v1-8k, etc. The LLM is only called when parsedatetime fails and the query has temporal intent, so costs stay near zero.

If TEMPORAL_LLM_MODEL is empty (the default), parsedatetime handles everything on its own. Requires the litellm package (pip install litellm or install Ogham with the appropriate extra).

Lifecycle hooks

Ogham hooks inject memory context at session start and preserve it across compaction. Install for your client:

ogham hooks install
Client What gets installed
Claude Code Hooks in ~/.claude/settings.json (recall on SessionStart/PostCompact, inscribe on PostToolUse/PreCompact)
Kiro Instructions for Hook UI (recall on Prompt Submit, inscribe on Agent Stop)
Codex, Cursor, others Project instruction file (CLAUDE.md, AGENTS.md, or .cursorrules)

Two commands, named after the Ogham stones:

  • recall -- read from the stone. Searches Ogham for memories relevant to your project and injects them as context. Fires at session start and after compaction.
  • inscribe -- carve into the stone. Captures meaningful tool activity as memories. Skips noise (ls, cat, git status) and only stores signal (commits, deploys, errors, config changes). Fires after tool use and before compaction. Secrets are masked before storing.

Smart filtering: Hooks don't capture everything. Routine commands (ls, pwd, git add) are skipped. Only signal events (errors, deployments, commits, config changes) are stored -- typically 20-30 memories per session instead of hundreds.

Secret masking: API keys, tokens, passwords, and JWTs are automatically replaced with ***MASKED*** before storing. The event is captured ("configured Stripe API key") but the actual secret never touches the database.

MCP tools

Memory operations

Tool Description Key parameters
store_memory Store a new memory with embedding content (required), source, tags[], auto_link
store_decision Store an architectural decision decision, reasoning, alternatives[], tags[]
update_memory Update content of existing memory memory_id, content, tags[]
delete_memory Delete a memory by ID memory_id
reinforce_memory Increase confidence score memory_id
contradict_memory Decrease confidence score memory_id

Search

Tool Description Key parameters
hybrid_search Combined semantic + full-text search (RRF) query, limit, tags[], graph_depth
list_recent List recent memories limit, profile
find_related Find memories related to a given one memory_id, limit

Knowledge graph

Tool Description Key parameters
link_unlinked Auto-link memories by embedding similarity threshold, limit
explore_knowledge Traverse the knowledge graph memory_id, depth, direction

Profiles

Tool Description Key parameters
switch_profile Switch active memory profile profile
current_profile Show active profile --
list_profiles List all profiles with counts --
set_profile_ttl Set auto-expiry for a profile profile, ttl_days

Import / export

Tool Description Key parameters
export_profile Export all memories in active profile format (json or markdown)
import_memories_tool Import memories with deduplication data, dedup_threshold

Maintenance

Tool Description Key parameters
re_embed_all Re-embed all memories (after switching providers) --
compress_old_memories Condense old inactive memories (full text to summary to tags) --
cleanup_expired Remove expired memories (TTL) --
health_check Check database and embedding connectivity --
get_stats Memory counts, profiles, activity --
get_cache_stats Embedding cache hit rates --

Skills

Ogham ships with three workflow skills in skills/ that wire up common MCP tool chains. Install them in Claude Code, Cursor, or any client that supports skills.

Skill Triggers on What it does
ogham-research "remember this", "store this finding", "save what we learned" Checks for duplicates via hybrid_search before storing. Auto-tags with a consistent scheme (type:decision, type:gotcha, etc.). Uses store_decision for architectural choices.
ogham-recall "what do I know about X", "find related", "context for this project" Chains hybrid_search, find_related, and explore_knowledge to surface connections. Bootstraps session context at project start.
ogham-maintain "memory stats", "clean up my memory", "export my brain" Runs health_check, get_stats, cleanup_expired, re_embed_all, link_unlinked. Warns before irreversible operations.

Skills call existing MCP tools -- they don't replace them. The MCP server must be connected for skills to work.

Install all three with npx:

npx skills add ogham-mcp/ogham-mcp

Or install a specific skill:

npx skills add ogham-mcp/ogham-mcp --skill ogham-recall

Manual install (copy from a local clone):

cp -r skills/ogham-research skills/ogham-recall skills/ogham-maintain ~/.claude/skills/

Scoring and condensing

Ogham goes beyond storing and retrieving. Three server-side features run automatically, no configuration needed.

Novelty detection. When you store a memory, Ogham checks how similar it is to what you already have. Redundant content gets a lower novelty score and ranks quieter in search results. You can still find it, but it won't push out more useful memories.

Content signal scoring. Memories that mention decisions, errors, architecture, or contain code blocks get a higher signal score. A debug session where you fixed a real bug ranks above a casual note about a meeting. The scoring is pure regex, no LLM involved.

Automatic condensing. Old memories that nobody accesses gradually shrink. Full text becomes a summary of key sentences, then a one-line description with tags. The original is always preserved and can be restored if the memory becomes relevant again. Run compress_old_memories manually or on a schedule. High-importance and frequently-accessed memories resist condensing.

Cross-encoder reranking

Optional FlashRank cross-encoder reranking for self-hosters who want better ranking precision. Adds ~300ms per search on CPU.

After Ogham's hybrid search returns candidates, FlashRank (ms-marco-MiniLM-L-12-v2, 21MB) rescores each result against the query using deeper token-level attention. The final score blends retrieval ranking with cross-encoder ranking.

BEAM benchmark impact: R@10 0.69 → 0.70, MRR +8pp. Biggest gain: temporal reasoning 0.84 → 0.98. Full results.

Install and enable:

pip install ogham-mcp[rerank]
# or: uv add ogham-mcp[rerank]

export RERANK_ENABLED=true
export RERANK_ALPHA=0.55   # 55% cross-encoder, 45% retrieval score

The model downloads on first use (~21MB). Self-hosters who want speed over precision leave it off (the default).

ONNX local embeddings

Run BGE-M3 locally with ONNX Runtime -- dense and sparse vectors in a single model pass, no API calls, no GPU required. Contributed by @ninthhousestudios.

The ONNX provider produces 1024-dim dense vectors plus neural sparse vectors. When sparse vectors are available, Ogham automatically uses three-signal Reciprocal Rank Fusion (dense + FTS + sparse) instead of the default two-signal path.

Install and configure:

pip install ogham-mcp[onnx]

# Download the model (~2.2GB)
ogham download-model bge-m3

export EMBEDDING_PROVIDER=onnx
export EMBEDDING_DIM=1024

Your database schema must use vector(1024) for the embedding column. Performance on CPU: ~0.3s per short text, ~10s for long documents (5K+ chars). RSS: ~4.3GB peak.

The ONNX provider is designed for self-hosters who want zero API costs. Cloud users should use Gemini or Voyage for lower latency.

Database setup

Ogham works with Supabase or vanilla PostgreSQL. Run the schema file that matches your setup:

File Use case
sql/schema.sql Supabase Cloud
sql/schema_selfhost_supabase.sql Self-hosted Supabase with RLS
sql/schema_postgres.sql Vanilla PostgreSQL / Neon (no RLS)

Supabase and Neon both include pgvector out of the box -- no extra setup needed. If you're self-hosting Postgres, you need PostgreSQL 15+ with the pgvector extension installed. We develop and test against PostgreSQL 17.

For Postgres, set DATABASE_BACKEND=postgres and DATABASE_URL=postgresql://... in your environment.

Upgrading from v0.4.x

If you already have an Ogham database, run the upgrade script to add temporal columns, halfvec compression, and lz4:

# Postgres / Neon (psql required)
./sql/upgrade.sh $DATABASE_URL

# Or run migrations individually
psql $DATABASE_URL -f sql/migrations/012_temporal_columns.sql
psql $DATABASE_URL -f sql/migrations/013_halfvec_compression.sql
psql $DATABASE_URL -f sql/migrations/014_lz4_toast_compression.sql
psql $DATABASE_URL -f sql/migrations/015_temporal_auto_extract.sql
psql $DATABASE_URL -f sql/migrations/016_sparse_embedding.sql
psql $DATABASE_URL -f sql/migrations/017_rrf_bm25.sql

# Supabase: paste each migration file into the SQL Editor

Migration 016 adds the sparse_embedding column for ONNX BGE-M3 sparse vectors. Migration 017 upgrades the search function to true Reciprocal Rank Fusion with length-normalised keyword scoring -- re-run your schema file to apply.

All migrations are idempotent -- safe to re-run. The upgrade script checks your pgvector version and skips halfvec if pgvector is below 0.7.0.

New installs don't need migrations -- the schema files already include everything.

Architecture

Ogham runs as an MCP server over stdio or SSE. Your AI client connects to it like any other MCP tool.

AI Client (Claude Code, Cursor, Kiro, OpenCode, ...)
    |
    | stdio (MCP protocol)
    |
Ogham MCP Server
    |
    | HTTPS (Supabase REST API) or direct connection (Postgres)
    |
PostgreSQL + pgvector

Memories are stored as rows with vector embeddings. Search combines pgvector cosine similarity with PostgreSQL full-text search using Reciprocal Rank Fusion (RRF) -- position-based, score-agnostic fusion that handles different score scales correctly. Optional FlashRank cross-encoder reranking adds a second pass for self-hosters. The Supabase backend uses postgrest-py directly (not the full Supabase SDK) for a lightweight dependency footprint.

The knowledge graph uses a memory_relationships table with recursive CTEs for traversal -- no separate graph database.

Documentation

Full docs and integration guides at ogham-mcp.dev.

Credits

Inspired by Nate B Jones and his work on persistent AI memory.

Named after Ogham, the ancient Irish alphabet carved into stone -- the original persistent memory.

License

MIT

Yorumlar (0)

Sonuc bulunamadi