Skip to main content
Glama
mgi-mind

mgi-mind

Official

Related Servers

Alternatives to mgi-mind

No user-submitted related servers found.

    Related Servers

    • F
      license
      Not graded
      quality
      B
      maintenance
      Gives AI agents their own long-term memory on the local machine: a graph of corrections, decisions, and behavioral experience that persists across sessions and is recalled by association rather than keyword. Runs fully offline in Rust, with background sleep consolidation, explanation of search results, and optional encryption — no cloud or LLM required.
      2
      -
    • A
      license
      A
      quality
      C
      maintenance
      Hermetic memory for AI agents — one Rust binary, one SQLite file, zero network. Recall returns evidence with provenance or abstains: no code path for making things up.
      22
      9
      AGPL 3.0
    • F
      license
      C
      quality
      A
      maintenance
      Memory for LLM agents: a local graph where the agent keeps decisions, tasks, hypotheses and reasoning between sessions, and gets a summary of what is open and overdue on every return to a project. Works with Claude Code, Codex, Cursor and opencode. Single Rust binary, no embedding model or LLM calls. Source-available (BSL 1.1), commercial use permitted.
      35
      89
      -
    • A
      license
      Not graded
      quality
      B
      maintenance
      Local memory for AI agents in one SQLite file on your own machine. Facts carry a validity window and a confidence score, so an outdated one is superseded and downweighted rather than deleted, and retrieval fuses BM25, vector similarity and a knowledge graph.
      4
      Business Source 1.1

    TDQS

    B3.2/5.0

    Scored across 44 tools

    Disambiguation3/5

    There are genuine clusters of near-synonymous tools: mind_search/mind_browse/mind_recall/mind_recall_all, mind_add/mind_ingest, and mind_outcome vs mind_procedure_outcome (the former explicitly subsumes the latter). Worse, ~15 deprecated aliases (mind_fact_add, mind_quarantine_list, mind_vault_get, mind_session_start, etc.) directly duplicate the newer action-based tools, so an agent can easily pick the wrong one. Descriptions are unusually careful, which keeps this from scoring lower, but the overlap is real.

    Naming Consistency4/5

    Nearly everything is snake_case under a shared mind_ prefix and follows a predictable verb_noun style (mind_search, mind_add, mind_learn). The main deviation is the new action-based noun-only tools (mind_fact, mind_library, mind_vault, mind_session, mind_quarantine) plus stateless names like mind_doctor and mind_context. Readable, but the mixed verb-style/noun-with-action convention is a minor inconsistency.

    Tool Count2/5

    44 tools is far beyond the reasonable band for this domain, and the count is inflated by ~15 deprecated tools that still appear in the surface alongside their replacements. Even the ~29 active tools is heavy for what is essentially a memory store. Deprecation flags help, but shipping both generations at once bloats the set.

    Completeness4/5

    The lifecycle is unusually thorough: add/ingest, semantic search, browse, cross-type recall, knowledge-graph facts, procedures, skills, pinned blocks, sessions, vault, quarantine, provenance, outcome signals, export/import, consolidate, restore, stats and health checks. The only minor gap is that there is no clear in-place edit/update path for an ordinary memory (append + invalidate/delete are the only mutations).

    Maintenance

    ActivityActive
    ResponsivenessLimited data