M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
COMPARE UP TO 4 SERVERS

Compare MCP servers

Compare scores, permissions, risks, and fit in one decision-focused table.

DimensionMisakaNet MCP ServerGit-backed failure memory for AI coding agents.NPM Sentinel MCP ServerAI-powered NPM package analysis MCP server
FMRS52 / 100 · D79 / 100 · B
Reliability6 / 2012 / 20
Security and permissions9 / 2016 / 20
Maintenance13 / 2018 / 20
Documentation13 / 2018 / 20
Setup experience11 / 2015 / 20
Best for
  • AI coding agents that need failure-recovery knowledge during task execution
  • Developers who want public debugging knowledge managed with Git and Markdown
  • Users who prefer zero-dependency, database-free deployment
  • Developers auditing NPM dependencies within AI workflows
  • Teams performing supply chain security assessments
  • Users of Claude Desktop, Cursor, or VS Code
Not for
  • A general-purpose memory system
  • An AI-agent runtime or framework
  • A vector database or RAG system
  • A skill marketplace
  • Projects outside the NPM ecosystem (e.g., pure Python/Go)
  • Environments without network access to deps.dev, OSV.dev, and the npm registry
  • Scenarios requiring maintenance by an official upstream vendor
Required permissions
  • Local stdio setup requires running Python 3 and accessing the MisakaNet source and lesson files
  • Remote MCP setup requires using the supplied Bearer token to access the remote endpoint
  • Contributing or submitting an intake report requires intentional submission of content
  • Network access to the NPM registry and external services (deps.dev, OSV.dev, OpenSSF, npms.io, GitHub)
  • Read access to workspace lockfiles (pnpm-lock.yaml, package-lock., yarn.lock) for cache invalidation
Risks and side effects
  • Lessons are community-contributed; review and sandbox commands before executing them
  • A retrieved fix may not apply to the current environment
  • Remote use involves sending searches to the service and handling an access token
  • Third-party READMEs and changelogs are untrusted external content; they are wrapped in tags with _meta flags but should still be handled cautiously
  • Depends on availability and accuracy of external services
  • Batch requests are capped at 25 packages to prevent registry enumeration; larger sets require batching
  • Third-party open source project, not officially maintained by NPM or Anthropic
Supported clientsCursor, Claude Code, CodexClaude Desktop, VS Code, Cursor
Tools019