The library
Everything we index — ranked by what works, never by stars.
forSalesMarketingHRFinanceLegalOpsProductEngineeringDataProductivitySupportsetup≤ plug & play≤ + a key≤ multi-tool
● works · ● untested / no effect · ● hurts — every rank is measured against a no-skill baseline
untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★0→untested★5,433→untested★2,474→untested★1,300→untested★292→untested★95→untested★365→untested★1,344→untested★95→untested★15→
Manage Stripe payments via AI agentsmcp_serverEngineeringOpsL2
mcp · Integrating Stripe payment operations into AI agents with permission control via Restricted API Keys.
Generate TypeScript clients from OpenAPI specsmcp_serverEngineeringL2
mcp · Backend developers automating type-safe API client generation from OpenAPI contracts.
Build production MCP servers with audit controlsmcp_serverEngineeringL3
mcp-server · MCP server developers needing production-hardened scaffolding with audit, metrics, and role-based access control.
Implement Figma designs in one shotmcp_serverEngineeringProductL2
figma-developer-mcp · Cursor and other AI IDEs that need clean, simplified design metadata to one-shot accurate component implementations.
Execute on-chain transactions without embedded keysmcp_serverEngineeringOpsL2
mcp-server · Agents executing on-chain operations that need non-custodial Phantom wallet integration without embedded keys.
Access SAP CAP data in AI agentsmcp_serverEngineeringOpsL2
mcp-server · SAP developers integrating CAP-CMS data access into AI agents.
Integrate any REST API with MCPmcp_serverEngineeringL2
openapi-mcp-server · Generic REST API integration when OpenAPI specs are available.
Bridge Claude with OpenAI Codexmcp_serverEngineeringL2
codex-mcp-server · Claude integration with OpenAI Codex for AI-powered code analysis.
Query pipeline and build statusmcp_serverEngineeringL2
mcp-server-circleci · CI/CD automation agents querying pipeline and build status.
Execute Postman API operationsmcp_serverEngineeringL2
postman-mcp-server · Use when agents need to dynamically test APIs or keep code synchronized with Postman collections.
Audit LLM traces and experiment metricsmcp_serverEngineeringL2
langsmith-mcp-server · Use when you need to audit LLM traces, pull prompt templates, or compare experiment performance.
Manage serverless Redis and queuesmcp_serverEngineeringL2
mcp-server · Use when agents need serverless Redis or message queue operations with automatic scaling.
Navigate code definitions semanticallymcp_serverEngineeringL2
lsp-mcp-server · Use as primary tool for code navigation over Grep — provides semantic accuracy across monorepos.
Fetch live Svelte documentationmcp_serverEngineeringL2
svelte-llm-mcp · Use when agents need latest Svelte/SvelteKit syntax without stale docs in training data.
Automate TouchDesigner parametersmcp_serverEngineeringL2
touchdesigner-mcp-server · Use when agents need to automate TouchDesigner parameter changes.
Execute remote commands over SSHmcp_serverEngineeringL2
ssh-mcp-server · Use when agents need execute remote commands with command security controls.
Manage hosting resources and deploymentsmcp_serverEngineeringL2
mcp-server · Use when agents need manage hosting resources via OAuth.
Automate GitHub Projects workflow statesmcp_serverEngineeringL2
ralph-hero-mcp-server · Managing GitHub issues through workflow states in Claude without shelling out to gh CLI.
Sync Qase test execution and defectsmcp_serverEngineeringL2
mcp-server · Syncing test execution and defect data from Qase into Claude workflows.
Run parallel multi-agent sessionsmcp_serverEngineeringL3
agent-orchestrator-mcp-server · Running multiple Claude agents in parallel with full lifecycle management and transcript persistence.
Observe and debug LLM tracesmcp_serverEngineeringL2
mcp-server · Observing and debugging LLM traces, managing prompt versions, and running scenario evaluations.
Manage infrastructure secrets via Dopplermcp_serverEngineeringL2
mcp-server · Managing infrastructure secrets from Claude when Doppler is your secrets source-of-truth.
Live edit Godot game scenes and scriptsmcp_serverEngineeringL3
godot-mcp-server · Live Godot editor manipulation from Claude with real-time scene/script sync and interactive dependency visualizer.
Manage infrastructure deployments via diviopsmcp_serverEngineeringL2
mcp-server · Managing infrastructure deployments from Claude when diviops is your IaC tool.
Automate GitLab projects and CI/CDmcp_serverEngineeringL2
gitlab-mcp-server · Managing GitLab projects and CI/CD from Claude when GitLab is your VCS and CI/CD platform.
Automate GitLab workflows cross-instancemcp_serverEngineeringL2
gitlab-mcp · When you need to automate GitLab workflows (MRs, pipelines, work items) with per-instance support and want schema filtered to what your tier actually supports.
Analyze Vue design system componentsmcp_serverEngineeringL2
musea-mcp-server · When you have a Vue design system and want LLM-driven component analysis, variant generation, and token lookup without leaving the MCP interface.
Control Platforma Desktop projectsmcp_serverEngineeringL3
pl-mcp-server · When you need agents to control a Platforma Desktop project directly (click, type, screenshot) rather than just calling APIs.
Search Docusaurus documentationmcp_serverEngineeringL2
docusaurus-plugin-mcp-server · When your docs are in Docusaurus and you want agents to search them with a plugin that's built-in to your static site.
Search versioned docs across npm PyPI GitHubmcp_serverEngineeringL2
docs-mcp-server · When you want version-specific, always-fresh documentation lookups across npm, PyPI, GitHub, and local files in a single tool.
Sync development context across Claude Cursor Windsurfmcp_serverEngineeringL2
mcp-server · When you hop between Claude Code, Cursor, and Windsurf and want shared session context (decisions, blockers, work summary) everywhere.
Compress webpages to token-optimized markdownmcp_serverEngineeringL2
compress-tokens · When you need to compress webpages to clean markdown for LLM context windows and want to save tokens on long articles.
Identify code security vulnerabilities deterministicallymcp_serverEngineeringL2
mcp · When you want instant, proof-backed security findings in your IDE or CI without noise and prefer deterministic rules to ML models.
Analyze large codebases with Gemini token windowmcp_serverEngineeringL2
gemini-mcp-tool · Analyzing large codebases and files using Google Gemini's massive token window.
Debug production incidents with observability datamcp_serverEngineeringOpsL2
mcp-server · Debugging production incidents by accessing real observability data from Last9.
Configure AI agents for enterprise search Gleanmcp_serverEngineeringOpsL2
configure-mcp-server · Configuring AI agents to access enterprise search via Glean's MCP servers.
Access GitLab projects with progressive disclosure controlsmcp_serverEngineeringL2
efficient-gitlab-mcp-server · Progressive-disclosure access to GitLab projects with read/write capabilities.
Build minimal MCP servers with typed toolsmcp_serverEngineeringL4
tiny-stdio-mcp-server · Building minimal MCP servers with typed tools and rich content helpers.
Provision manage GPU compute infrastructure on demandmcp_serverEngineeringOpsL3
mcp-server · On-demand GPU clusters for AI inference or fine-tuning without Kubernetes complexity.
Generate typed CLI commands from MCP toolsmcp_serverEngineeringL3
mcp · Bridge MCP servers into developer CLIs with automatic argument coercion and caching.
Search curated resources from 8500 awesome listsmcp_serverEngineeringL2
context-awesome · Agents needing high-quality vetted resource recommendations instead of general web search.
Format code and create pull requestcommandEngineeringL1
pr · One-command workflow to atomically format, stage, and push a PR without manual git steps.
Answer code navigation and impact questionscommandEngineeringL1
Q · MCP-first code intelligence queries when you need call/def/ref across the whole codebase without manual grep.
Run ChunkHound QA scriptcommandEngineeringL1
qa · Verifying semantic and regex search implementations work correctly across all file types and languages without blocking or pagination bugs.
Review uncommitted code changescommandEngineeringL1
cr · Catching language-specific bugs and design issues before commit by applying both project-local and generic review profiles to changed code.
Run CI checks locallycommandEngineeringL1
ci · Testing a Haskell-to-Rocq compiler change locally before pushing by running the exact GitHub Actions workflow steps without Docker.
Manage agent-view sessionscommandEngineeringL1
av · Programmatically launching and controlling multiple AI agent sessions without TUI overhead when orchestrating batch work or CI/CD-driven agent tasks.
Create clean PR with planning artifactscommandEngineeringL1
pr · Separating code review (clean diff) from planning artifact archiving, allowing reviewers to focus on implementation while preserving design rationale as write-once history.
Configure recursive language model modecommandEngineeringL1
rlm · Deep codebase analysis on large repositories requiring multi-level reasoning without manual context pruning, routing models and tool access by depth/effort.
Run MySQL queries on Docker databasecommandDataEngineeringL1
db · Quick ad-hoc database queries during development when Docker Compose is the deployment model and CLI access is faster than GUI tools.