Skip to main content
Glama
SVerITG

Metis

by SVerITG
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

Related Servers

Alternatives to Metis

No user-submitted related servers found.

    Related Servers

    • rag-pipeline

      jansc4
      F
      license
      A
      quality
      C
      maintenance
      Enables Claude to search a local hybrid retrieval index of research papers and ingest new PDFs, providing research-paper memory queryable directly through natural language.
      2
      -

    • Personal Research Assistant

      chakri01
      F
      license
      Not graded
      quality
      D
      maintenance
      Enables semantic search and conversational querying across a personal research library of PDFs, DOCX, and other documents using a vector database. It provides tools for document summarization, finding related papers, and high-accuracy retrieval for AI clients like Claude Desktop.
      -

    • paperqa-mcp-server

      menyoung
      A
      license
      A
      quality
      D
      maintenance
      Give Claude the ability to read, search, and synthesize across your entire PDF library. Built on PaperQA2.
      2
      2
      MIT

    • ScholarScout

      amnanazir101-maker
      A
      license
      Not graded
      quality
      C
      maintenance
      Enables Claude Code and Claude Desktop to run multi-agent literature research: searching arXiv and OpenAlex for papers and producing cited literature reviews. Every cited statement is fact-checked against the paper it cites, so references cannot be fabricated.
      MIT

    • Torus

      sekhar14
      A
      license
      B
      quality
      A
      maintenance
      Gives agents a persistent local knowledge graph of companies, sourced claims and investment theses, so new research builds on what was already concluded and every fact keeps its citation. It also watches SEC filings and news, raises alerts on the theses they affect, and supports onboarding, challenge, memo and review workflows.
      4
      45
      1
      Apache 2.0

    • Agentic Coach Builder

      kaushalag29
      F
      license
      Not graded
      quality
      B
      maintenance
      Enables Claude Code and Cowork to act as a personal coach grounded in a locally indexed library of YouTube talks, websites and PDF books, answering with exact-second, paragraph or page citations and recalling the user's goals and decisions per Project. Search, memory and the usage log run on the user's own machine, and changing Sources rebuilds the knowledge base without any fine-tuning.
      -
    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

    B3.1/5.0

    Scored across 283 tools

    Disambiguation2/5

    Several tool clusters genuinely overlap: metis_doctor vs verify_install, find_connections vs cross_pollinate, search_literature vs search_literature_extended, propose_skill_improvement/approve_proposal vs draft_self_improvement_proposal/apply_proposal (two parallel proposal systems), and a memory-write family of remember / add_memory_entry / store_episodic_memory / store_semantic_memory / store_procedural_memory / capture_observation. Descriptions work hard to cross-reference each other ('for X use Y; for Z use W'), which rescues many cases, but with 283 tools the boundaries still blur and misselection is likely.

    Naming Consistency3/5

    snake_case is used uniformly, which is good, but the verb_noun convention is broken often: recall, remember, metis_doctor, kg_paths, kg_community, focus_brief, focus_safe, reading_stack_show, nature_briefing, viz_library_overview, dhis2_query/dhis2_metadata mix noun-first, question-style, and prefixed forms. Readable overall, but no single predictable pattern an agent can rely on.

    Tool Count1/5

    283 tools is an extreme mismatch for any single agent context, far past the 15-tool comfort zone. The presence of find_tools / load_tool_group / list_tool_groups shows the team knows this and has built lazy loading, but the declared surface is still enormous and forces every caller through a discovery step.

    Completeness4/5

    Coverage is unusually broad: full CRUD for projects, tasks, memories, library items, ideas, journal, contacts; plus verification, backups, DHIS2, course building, document/poster generation, and visualization tracking. Gaps exist (e.g. no update for jobs* beyond archive/unarchive) but almost any workflow has a path.

    Maintenance

    ActivityActive
    ResponsivenessUnresponsive

    Resources