Skip to content

README.md

Latest commit

 

History

History
67 lines (45 loc) · 2.46 KB

README.md

File metadata and controls

67 lines (45 loc) · 2.46 KB

DeepSource Skills

A collection of agent skills for working with DeepSource products. Skills are packaged instructions and scripts that extend AI coding agents with DeepSource capabilities.

Available Skills

sentinel-api

Scan repositories for security vulnerabilities, leaked secrets, and dependency issues using the Sentinel API, and auto-fix detected issues.

  • Use when: you want to run a security scan on a repository or codebase using Sentinel
  • Use when: you want to analyze code changes or pull requests for vulnerabilities
  • Use when: you want to auto-fix security issues detected by Sentinel
  • Includes: sync and polling scripts for automated workflows, full API reference

deepsource

Retrieve code review results from DeepSource using the DeepSource CLI — issues, vulnerabilities, report cards, and analysis runs.

  • Use when: you want to list code review issues, bugs, or security findings for a repository
  • Use when: you want to check dependency vulnerabilities or CVE status
  • Use when: you want to get code review grades, coverage metrics, or report cards
  • Use when: you want to view analysis run history

Installation

Install all skills:

npx skills add DeepSourceCorp/skills

Install a specific skill:

npx skills add DeepSourceCorp/skills -s sentinel-api
npx skills add DeepSourceCorp/skills -s deepsource

Skills work with 18+ AI agents including Claude Code, GitHub Copilot, Cursor, Cline, Codex, and more.

Usage

Once installed, skills are automatically available to your AI agent. Just ask naturally:

  • "Scan this repo for security vulnerabilities using Sentinel"
  • "Run Sentinel on my latest changes and fix any issues"
  • "Upload this repository to Sentinel and analyze it"
  • "Show me the critical code issues on DeepSource"
  • "Check for dependency vulnerabilities on DeepSource"
  • "What's the DeepSource report card for this pull request?"

Skill Structure

Each skill follows the Agent Skills specification:

skill-name/
├── SKILL.md           # Instructions and metadata (required)
├── scripts/           # Executable helper scripts (optional)
├── references/        # Detailed reference documentation (optional)
└── assets/            # Static resources (optional)

License

MIT