도구

Helps hiring managers design structured interviews for a role and level: 4-6 competencies, rounds, a question bank with follow-ups and answer criteria, 1-4 scorecards, a take-home task, debriefs and bias safeguards. Prompt-only; startup-focused, pairs with the startup-context skill.

Helps hiring managers write or rewrite job descriptions: first clarify what problem the role solves, then follow a fixed structure, cut buzzwords, inflated requirements and gendered wording, and prompt for a salary range. Startup-focused; best with company info from the startup-context skill.

Helps HR and founders write outreach to passive candidates: LinkedIn messages, cold emails and referral requests, drafting a first message and 2-3 follow-ups around personalization, relevance and a call to action. Drafts only, never sends; startup-focused.

50 skills for founders covering fundraising (pitch materials, investor research, data rooms), sales outreach, PRDs, hiring (JDs, interview questions) and legal docs (privacy policy, terms of service, contract review). Set up startup-context once; have a lawyer check legal output.

Apache Doris's official MCP: gives agents read-only access to explore table metadata, run and explain SQL, diagnose slow queries, and check cluster status, load jobs and data governance info. Needs Python 3.12+ and Doris 2.0 or later.

Railway's official plugin, bundling the use-railway skill and hosted Railway MCP so agents can create projects, deploy services, set environment variables and databases, and check logs and metrics to debug deployments. Needs a Railway account.

ComfyUI's official MCP: lets agents drive local ComfyUI image and video workflows, list installed nodes, models and templates, submit jobs and fetch results, and start or stop ComfyUI. Needs a machine that can run ComfyUI (usually a GPU) and comfy-cli; partner models consume Comfy credits.

Vercel's official plugin: a knowledge graph of the Vercel ecosystem, dozens of skills, expert subagents for deployment and performance, and commands for deploys, env vars and project status, covering Next.js, the AI SDK, storage and CI/CD. Needs Node.js 18+ and Bun; telemetry is opt-out.

QuantConnect's official CLI drives the open-source LEAN engine locally or in the cloud to download data, backtest and optimize quant strategies and deploy them live. Requires a paid-tier QuantConnect organization; strategies in Python or C#; local runs need Docker. Get expert review.

Hugging Face's official MCP server: lets agents search models, datasets and Spaces on the Hub and call thousands of Gradio AI apps directly. Hosted remotely; sign in with your Hugging Face account or authorize with a token.

Query and manage Microsoft 365 and Azure tenants in natural language via Microsoft Graph: create users and security groups, check devices and Conditional Access policies and view cloud costs. Needs a work account with the right permissions; changes apply immediately.

Grafana Labs' official skills: dashboards and alerting, PromQL, Loki, Tempo, Mimir and Pyroscope queries, Prometheus cardinality troubleshooting, Grafana Cloud management and cost optimization, k6 load testing and plugin development. Some skills need Grafana Cloud.

Lets agents create, update and resolve incidents in ServiceNow, manage the service catalog, change requests and knowledge articles, and load only the tools each role needs. Needs a ServiceNow account and Python 3.11+ (from source); community-maintained, last updated Oct 2025.

Alibaba Cloud's official agent skills: ECS diagnosis and DR, RDS/PolarDB database checks, PAI training and EAS deployment, ACK clusters, RAM permissions, OSS and network troubleshooting, and Model Studio. Needs an AccessKey or aliyun CLI credentials.

Puts Claude in your incident channel: builds triage runbooks from past incidents, investigates alerts read-only and gives an evidence-backed first diagnosis while humans decide fixes, plus handoffs and status reports. Anthropic reference code, no longer maintained; needs Claude Team or Enterprise.

Firecrawl's official plugin: brings the Firecrawl CLI in as skills so Claude Code can search the web, scrape pages to Markdown, map site links or crawl whole sites. Install firecrawl-cli and set an API key first; self-hosting supported.

MinerU's official MCP: parses PDF, Word, PowerPoint and images into Markdown via the mineru.net cloud API, keeping tables and formulas. The keyless Flash mode works out of the box; add a token for higher quotas and more formats.

Comet's official Opik MCP: query traces and spans of LLM apps in chat, log scores and comments, and manage datasets, experiments and prompt versions. Works with Opik Cloud (API key or browser sign-in) and self-hosted open-source Opik.

Wraps Shapely, GeoPandas, PyProj, Rasterio, PySAL and other GIS libraries so agents can run geometry and overlay analysis, reproject coordinates, process rasters (e.g. NDVI), do spatial statistics and make maps, and fetch boundaries, land cover and satellite imagery. Needs Python 3.10+ and uv.

MongoDB's official plugin: runs the MongoDB MCP locally via npx and adds skills for query generation, index tuning, schema design, and full-text and vector search. Connects to self-hosted or Atlas databases with a connection string.

MongoDB's official skill library covering document model design, driver connection setup, query optimization, natural-language query generation, Atlas full-text and vector search, and stream processing. Works alongside the MongoDB MCP Server.

Official Alibaba Cloud MCP: query logs, metrics and traces from SLS, CloudMonitor and ARMS, with time-series comparison and natural-language-to-SQL. Needs an AccessKey; some AI tools are billed per call. The install uses the Python version (main branch is now a Go binary).

HeyGen's official CLI, built for agents like Claude Code and Codex: generate avatar presenter videos from a prompt or template, translate and dub videos, and synthesize speech, with JSON output. Good for batch training and product videos. Needs a HeyGen API key (usage-billed) or a Pro/Max plan.

Microsoft Translator's translation quality tool: GPT scores each sentence using MQM error categories or a direct score; --list_mqm_errors shows error types and severity. Needs line-aligned source and target files, Python and an OpenAI or Azure OpenAI key (paid); a translator should still review.