Skip to main content
Glama
Sign In
mcp-memory-graph

mcp-memory-graph

by YonasValentin
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Hybrid

Related Servers

Alternatives to mcp-memory-graph

No user-submitted related servers found.

    Related Servers

    • coldframe

      coldzero94
      A
      license
      Not graded
      quality
      B
      maintenance
      Local-first memory for your AI agent. One SQLite file you own — offline, no API key. Plugs straight into Claude Code.
      Apache 2.0

    • Yggdrasil

      VonderVuflya
      A
      license
      A
      quality
      B
      maintenance
      Durable, local-first memory for AI coding agents over MCP — zero-dependency (pure Python + SQLite/FTS5), curated and semantically de-duped. Works with Claude Code, Codex and any MCP host, and you own the data as plain rows.
      6
      67 PyPI
      22
      AGPL 3.0

    • Synapse

      alexsh88
      F
      license
      Not graded
      quality
      C
      maintenance
      A self-hosted, temporal knowledge-graph memory for AI coding agents — shared across projects, queryable across time, written and read via MCP by any Claude Code session.
      -

    • Agent Memory

      heidrickla
      A
      license
      Not graded
      quality
      B
      maintenance
      Enables Claude Code, Codex and other MCP agents to share one bounded, persistent memory across sessions, so a session starts from what earlier sessions learned instead of re-reading the repository. Storage and retrieval run locally on SQLite full-text search with per-client tokens and scopes, and optional local model enrichment, without any paid model in the path.
      1
      MIT

    • SuperMemory

      Yuziki1227
      A
      license
      Not graded
      quality
      C
      maintenance
      Enables Claude Code and Codex to write, update, and retrieve durable long-term memories that survive session boundaries and context compaction. All recall runs locally against a single SQLite file using hybrid BM25, embedding, and recency ranking cut to an explicit token budget, with no cloud dependency.
      MIT
    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A3.7/5.0

    Scored across 51 tools

    Disambiguation4/5

    Most memory_* tools map to distinct actions or resources, and richly detailed descriptions separate overlapping areas like semantic search, graph traversal, and versioning. A few pairs remain confusable—memory_versions vs memory_history and memory_delete vs memory_forget—so it is not quite a perfect 5.

    Naming Consistency4/5

    The dominant memory_<verb> pattern is clear and mostly consistent, but many tool names are noun-style rather than verb-style (memory_manifest, memory_questions, memory_health, memory_graph). There are also prefix exceptions like vault_sync and core_memory_* mixed with memory_export_vault.

    Tool Count2/5

    51 tools is well above the heavy threshold. Many niche capabilities—webhook management, expertise profiles, session state, dataset export, health reporting—are single-purpose subfeatures that could have been grouped, making the surface unnecessarily large for an agent to navigate.

    Completeness5/5

    The server covers the full memory lifecycle: create, read, update, delete, tombstone, ingest, import, export, versioning with restore/diff, graph entities and communities, vault sync, and system health/governance. The domain is thoroughly covered with no obvious dead ends.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Resources