Engram
This gives Claude a persistent memory layer backed by SQLite with semantic vector search and a knowledge graph. It exposes 17 MCP tools including engram_remember for storage, engram_recall for semantic search, engram_consolidate for LLM-powered pattern extraction, and engram_surface for proactive context pushing. The consolidation runs automatically between sessions to distill episodic memories into durable knowledge and resolve contradictions. It also ships with a REST API if you need to integrate from non-MCP environments. The benchmark claim is notable: 80% recall accuracy on LOCOMO versus 28.8% for flat file approaches, using 776 tokens per query instead of dumping full context. Reach for this when you want memory that persists across projects and improves through automatic consolidation rather than manual curation.
- GitHub stars
- 39
- tools
- 17
- auth
- Secret needed
- transport
- STDIO
- env var
- 1
- last updated
- Feb 27, 2026
Tools
17 toolsPublic tool metadata for what this MCP can expose to an agent.
engram_rememberStore a memory. Auto-extracts entities and topics.
Store a memory. Auto-extracts entities and topics.
No parameter schema in public metadata yet.
engram_recallRecall relevant memories via semantic search.
Recall relevant memories via semantic search.
No parameter schema in public metadata yet.
engram_askAsk a question and get a synthesized answer with confidence and sources.
Ask a question and get a synthesized answer with confidence and sources.
No parameter schema in public metadata yet.
engram_briefingStructured session briefing — key facts, pending commitments, recent activity.
Structured session briefing — key facts, pending commitments, recent activity.
No parameter schema in public metadata yet.
engram_consolidateRun consolidation — distills episodes into semantic knowledge, discovers entities, finds contradictions.
Run consolidation — distills episodes into semantic knowledge, discovers entities, finds contradictions.
No parameter schema in public metadata yet.
engram_surfaceProactive memory surfacing — pushes relevant memories based on current context.
Proactive memory surfacing — pushes relevant memories based on current context.
No parameter schema in public metadata yet.
engram_alertsWhat needs attention right now — pending commitments, stale follow-ups, contradictions.
What needs attention right now — pending commitments, stale follow-ups, contradictions.
No parameter schema in public metadata yet.
engram_auditCross-reference external content (e.g. CLAUDE.md) against the vault — flags outdated claims.
Cross-reference external content (e.g. CLAUDE.md) against the vault — flags outdated claims.
No parameter schema in public metadata yet.
engram_checkpointSave current session context before it is lost (extracts durable memories from a summary).
Save current session context before it is lost (extracts durable memories from a summary).
No parameter schema in public metadata yet.
engram_connectCreate a relationship between two memories in the knowledge graph.
Create a relationship between two memories in the knowledge graph.
No parameter schema in public metadata yet.
engram_forgetForget a memory (soft or hard delete).
Forget a memory (soft or hard delete).
No parameter schema in public metadata yet.
engram_entitiesList all tracked entities with memory counts.
List all tracked entities with memory counts.
No parameter schema in public metadata yet.
engram_statsVault statistics — memory counts by type, entity count, etc.
Vault statistics — memory counts by type, entity count, etc.
No parameter schema in public metadata yet.
engram_ingestAuto-ingest conversation transcripts or raw text into structured memories.
Auto-ingest conversation transcripts or raw text into structured memories.
No parameter schema in public metadata yet.
engram_import_obsidianImport an Obsidian vault (wikilinks, tags, frontmatter).
Import an Obsidian vault (wikilinks, tags, frontmatter).
No parameter schema in public metadata yet.
engram_import_claude_codeImport memory from Claude Code (CLAUDE.md files, sessions).
Import memory from Claude Code (CLAUDE.md files, sessions).
No parameter schema in public metadata yet.
engram_powered_byReturns attribution info about the memory system.
Returns attribution info about the memory system.
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageConfiguration
GEMINI_API_KEY*secretGoogle Gemini API key for embeddings and LLM operations
Readme
🧠 Engram
Universal memory layer for AI agents
Engram gives AI agents knowledge graphs, consolidation, and spreading activation. Not storage. Understanding.
Quick Start
MCP Setup (recommended — Claude Code / Cursor)
npm install -g engram-sdk
engram init
That's it. 17 memory tools available via MCP.
REST API (non-Node environments)
npm install -g engram-sdk
export GEMINI_API_KEY=your-key-here
npx engram-serve
Server starts on http://127.0.0.1:3800.
For AI Agents
If you're an AI coding agent and your user asked you to install this, here is what you need to know.
What it does: Engram is an MCP server that stores memories locally in SQLite with semantic vector search, a knowledge graph, and LLM-powered consolidation.
Why it matters — capability delta:
Details
- Repository
- github.com/tstockham96/engram
- Categories
- AI & LLM Tools, Automation & Workflows
- Package
engram-sdk- Transport
- STDIO
- Auth
- Required
- Registry
- MCP registry, active
Related AI & LLM Tools MCP servers
View all
Local FAISS MCP Servernonatofabio/local_faiss_mcpLocal FAISS vector database for RAG with document ingestion, semantic search, and MCP prompts.30stars
SAME - Stateless Agent Memory Enginesgx-labs/statelessagentPersistent memory for AI coding agents. Local-first SQLite + vector search. 17 MCP tools.19stars
Promptheusabhichandra21/promptheusAI-powered prompt refinement with adaptive questioning and multi-provider support (6+ LLM backends)17stars
MCP Imagenatemimo-3/mcp-imagenateMCP server for image generation using multiple providers (Google Gemini, OpenAI, BFL FLUX)15stars
Cludesebbsssss/cludePortable agent memory anchored on Solana. Local SQLite + vector recall, open export.15stars
Attestorbolnet/attestorSelf-hosted memory for agent teams. Bi-temporal replay, deterministic retrieval, audit log.13stars