JetBrains/skills

Curated agent skills collection verified by JetBrains

What it solves

Finding reliable, ready-to-use agent skills is hard. Skills are scattered across many repositories with inconsistent quality and security. This catalog centralizes a JetBrains-filtered and verified snapshot of upstream skills, making them easy to discover and reuse, while also providing automated security scanning to catch problems early.

How it works

The repository collects skills from 15 upstream sources into a flat directory layout, each skill in its own top-level directory. Each SKILL.md file includes a metadata.source link pointing back to the original upstream skill. The catalog is maintained with attribution rules: the Author column uses explicit skill front matter if present, otherwise it falls back to the upstream maintainer or organization. A security scanning pipeline uses Cisco's skill-scanner. You can run it manually via a script, and CI workflows automatically scan on push, pull request, and manual dispatch. Push/PR runs only scan changed directories and fail on error-level findings in those changes. A full-repo audit runs on manual dispatch, uploading the report artifact without blocking on inherited upstream findings. The workflow can also push SARIF results to GitHub Code Scanning if enabled.

Who it’s for

Developers, AI agent builders, and teams that want to integrate reliable agent skills into their workflows. It's especially useful for those using JetBrains IDEs, though the skills themselves are generic and come from various ecosystems (Anthropic, OpenAI, Google, Vercel, Kotlin, etc.). The security scanning makes it suitable for organizations that need to vet third-party skills before adoption.

Highlights

  • 129 skills from 15 upstream repositories, covering coding, testing, design, cloud, and more.
  • Every skill retains a metadata.source link to its original upstream path for traceability.
  • Drop-in flat directory layout for easy discovery and reuse.
  • Automated security scanning with Cisco's skill-scanner: manual script, CI on push/PR (scans changed dirs), and full-repo audit on manual dispatch.
  • CI fails on error-level findings in changed skills; SARIF upload to GitHub Code Scanning is best-effort.
  • Attribution falls back to upstream maintainer/organization when skill front matter is absent.

Related

  • Project
    Tejas-TA/predikit

    A library that wraps scikit-learn and XGBoost models as LLM-callable tools with auto-generated JSON schemas and Pydantic validation.

  • Project
    yeyu2/Youtube_demos

    A repository filled with code examples for AI projects, each linked to a YouTube video. It covers multi-agent frameworks, realtime voice/multimodal assistants, RAG pipelines, and local model deployment, giving developers ready-made starting points.

  • Project
    buildingjoshbetter/TrueMemory

    A local-first long-term memory system for AI agents that automatically captures and updates user preferences and facts across sessions to eliminate AI amnesia.

  • Project
    fancyboi999/goofish-cli

    goofish-cli is a Python CLI that automates Xianyu (Chinese second‑hand marketplace) operations—publishing items, managing listings, sending/receiving messages, searching, etc. It outputs data in JSON/YAML/table formats, includes rate‑limiting and risk‑guard checks, and registers as an MCP server so AI agents (Claude Code, Cursor, OpenClaw) can call its commands as tools. Five Claude skill bundles are shipped for zero‑shot‑free agent use. The package is pip‑installable, open‑source under Apache 2.0, and comes with unit tests and documentation.