Skip to main content
Glama
Filipe-Soares-de-Almeida

MemAI

by Filipe-Soares-de-Almeida
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

Related Servers

Alternatives to MemAI

No user-submitted related servers found.

    Related Servers

    • NeuralVaultCore

      getobyte
      A
      license
      Not graded
      quality
      C
      maintenance
      An MCP server that provides persistent long-term memory for AI agents via local SQLite storage with low token overhead, enabling memory storage, retrieval, and management across sessions.
      1
      MIT

    • Agentic Memory

      ArkaAiAdmin
      A
      license
      A
      quality
      B
      maintenance
      MCP server for persistent, cross-session, local-first memory for AI agents, storing memories as Markdown files with SQLite indexing for hybrid search.
      24
      Apache 2.0

    • agenticmemory

      jyswee
      F
      license
      Not graded
      quality
      C
      maintenance
      Persistent memory MCP server for AI agents that stores, recalls, and searches conversation history, key-value context, and long-term entries across sessions with semantic search and FIFO queues.
      256 npm
      1
      -

    • heropen

      Koradji77
      A
      license
      Not graded
      quality
      A
      maintenance
      A local memory server for AI agents that stores and retrieves information via MCP, keeping all data in SQLite on your machine.
      204 PyPI
      1
      Apache 2.0
    Deploy Server
    Context All Your Agents Can ShareAgents share what they learn, hand off work, and hold it when you switch models. Start with 5 GB free.fulcradynamics.comAd

    TDQS

    A3.6/5.0

    Scored across 43 tools

    Disambiguation3/5

    The core write verbs (note, checkpoint, anti_pattern, reasoning) and the task_/diagram_ families are well separated, but the read surface overlaps heavily: recall is explicitly 'sugar for search(type="note")', and list_recent, list_by_domain, must_read, and pulse all return 'most recent memories' with only subtle scope differences. dedup_scan and optimize_scan also cross into the same 'find suspect memories' territory. Descriptions do a lot of work to disambiguate, but an agent still faces real misselection risk.

    Naming Consistency3/5

    Within domains the prefixes are consistent (task_add/task_read/task_note, diagram_node/diagram_edge), but across the server conventions clash: bare nouns (note, checkpoint, task, diagram), bare verbs (search, forget, pulse), verb_noun (get_memory, list_domains), and noun_verb (task_add, diagram_link) all coexist. Generally readable, but not a single predictable pattern.

    Tool Count2/5

    43 tools is a very heavy surface for a memory store and pushes well past the 'too many' threshold. Several clusters could be collapsed (four optimize_* tools, four separate diagram-edge/link/jump verbs, near-duplicate read tools), so the count feels inflated relative to what an agent can reliably choose among.

    Completeness5/5

    Coverage is genuinely thorough: create/read/update/delete across multiple memory types, soft vs hard delete, relations, edit history/versioning, confidence, domain and project management, plus a full curation workflow. Almost no obvious lifecycle gap remains for a memory-management domain.

    Maintenance

    ActivityActive
    ResponsivenessNo issues

    Resources