1. Conduid
  2. RAG
  3. Engram MCP
MCP server · RAG

Engram MCP

MCP server in Rust for AI agent persistent memory: branch-aware session handoffs, local ONNX embeddings, SQLite-backed semantic search.

Unclaimed rag
34Low

Scored 12 days ago · breakdown

About Engram MCP

Engram MCP is an MCP server published by edg-l in the RAG category: mCP server in Rust for AI agent persistent memory: branch-aware session handoffs, local ONNX embeddings, SQLite-backed semantic search.

The repository has 18 stars and 4 forks, with the last commit 4 days ago. Recent commit activity is the strongest signal that reported issues will get attention.

Install

From source
git clone https://github.com/edg-l/engram-mcp

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Tools (24)

adr_createCreate a numbered, Nygard-style ADR; optionally supersede an existing ADR

Parameters not published.

adr_exportExport ADRs to Nygard-style Markdown files (dry run by default)

Parameters not published.

adr_listList the project's ADRs, optionally filtered by status

Parameters not published.

adr_showShow a single ADR by number

Parameters not published.

adr_update_statusAdvance an ADR through its lifecycle (proposed/accepted/deprecated/rejected)

Parameters not published.

handoff_createCapture a session handoff with structured sections (summary, decisions, todos, blockers, mental_model, next_steps, notes)

Parameters not published.

handoff_resumeRetrieve top sections from recent handoffs on the current branch, plus linked memories

Parameters not published.

handoff_searchSearch handoff sections by content; filter by branch or section name

Parameters not published.

memory_contextLoad relevant memories for a task (hierarchical retrieval via clusters)

Parameters not published.

memory_dedupFind and merge duplicate memories (dry run by default)

Parameters not published.

Show all 24 tools
memory_deleteRemove a memory and its relationships

Parameters not published.

memory_delete_batchDelete multiple memories by ID

Parameters not published.

memory_exportExport project memories to JSON

Parameters not published.

memory_graphTraverse relationship graph from a root memory

Parameters not published.

memory_importImport from JSON (merge or replace modes)

Parameters not published.

memory_linkCreate typed relationships between memories

Parameters not published.

memory_projectsList every project in the store with memory, handoff, and ADR counts

Parameters not published.

memory_promotePromote a branch-local memory to global scope

Parameters not published.

memory_pruneRemove low-relevance memories (dry run by default)

Parameters not published.

memory_querySemantic search with hybrid scoring, pagination, branch filtering

Parameters not published.

memory_statsProject statistics (counts, types, pinned, global, clusters)

Parameters not published.

memory_storeStore a memory with embedding, auto-dedup, auto-cluster

Parameters not published.

memory_store_batchStore up to 100 memories atomically

Parameters not published.

memory_updateUpdate content, tags, importance, pinned status

Parameters not published.

Ask AI

✦

Ask AI about Engram MCP

Powered by Claude · Grounded in docs

I know everything about Engram MCP. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

6 checks have not been run yet.

Questions

About Engram MCP

How do I install Engram MCP?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is Engram MCP safe to use with an AI agent?

Its trust score is 34 out of 100 (low). Conduid hasn't run static security checks on this repository yet, so review the source yourself before granting it credentials. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Engram MCP still maintained?

Yes — the last commit was 4 days ago. The repository has 18 stars and 4 forks.