Skip to main content
Glama
cachly-dev

Cachly — AI Cognitive Brain

Related Servers

Alternatives to Cachly — AI Cognitive Brain

No user-submitted related servers found.

    Related Servers

    • A
      license
      A
      quality
      C
      maintenance
      Persistent cloud memory for AI coding assistants. 28 MCP tools for semantic search, auto-learning, task tracking, correction patterns, knowledge graphs, and session replay across Claude Code, Cursor, Windsurf, Cline, and any MCP client. Encrypted at rest. Team shared memory with author attribution.
      35
      415 npm
      7
      MIT
    • A
      license
      A
      quality
      D
      maintenance
      Persistent memory API for AI agents — store, recall, and inject semantically-searchable context across sessions. EU-hosted, GDPR-compliant. Supports Claude, Cursor, Cline, and any MCP-compatible client.
      4
      2
      MIT
    • A
      license
      A
      quality
      C
      maintenance
      AI memory that works like yours. Neural pathway architecture gives your AI persistent recall, contextual awareness, and cross-session continuity the way human memory actually works. 14 MCP tools. Works with Claude Code, Cursor, Windsurf, and any MCP client. $50/mo unlimited.
      15
      MIT
    • A
      license
      Not graded
      quality
      A
      maintenance
      Persistent memory for any AI assistant. Zero token cost until recall. Stores memories in local SQLite, ranks by 6-factor scoring, returns results 79% smaller than JSON. Works with Claude, ChatGPT, Grok, Cursor, Windsurf, and any MCP client.
      49
      Apache 2.0
    • F
      license
      Not graded
      quality
      B
      maintenance
      Persistent context layer for AI coding assistants. Sync memory files from Claude Code to ChatGPT, Gemini, Cursor, Windsurf, and Copilot via a native MCP endpoint. Bearer token auth, JSON-RPC 2.0 over stateless HTTP. Free tier available.
      -

    TDQS

    A3.8/5.0

    Scored across 122 tools

    Disambiguation3/5

    Tools are grouped with clear prefixes (cache_, brain_, team_, global_, syndicate_), but within each group there are many tools with overlapping scopes. For example, recall_best_solution, smart_recall, team_recall, global_recall, and brain_search all retrieve knowledge with subtle differences. Detailed descriptions help but the volume forces agents to read carefully to avoid misselection.

    Naming Consistency4/5

    Overall naming follows a consistent snake_case verb_noun pattern with descriptive prefixes (e.g., cache_get, brain_conflicts, team_assign_role). A few tools like trace_dependency and compact_recover lack prefixes, and some use acronyms (madc_deliberate, cls_ingest), but the majority is predictable and well-organized.

    Tool Count2/5

    122 tools is far beyond typical MCP servers (3-15). Even for a comprehensive cognitive brain system, this number risks overwhelming agents and makes the set hard to navigate. Many tools could be consolidated (e.g., multiple session utilities, multiple recall variants with overlapping functionality).

    Completeness5/5

    The tool set is extremely comprehensive, covering every aspect of the declared purpose: cache CRUD and monitoring, lesson lifecycle (learn, recall, confirm, resolve, consolidate), team management, global knowledge commons, federation, session management, roadmap, and knowledge graph inspection. No obvious gaps exist for the stated domain.

    Maintenance

    ActivityActive
    ResponsivenessResponsive