setup_skill
- TypeScript
10.5k
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill everyinc/compound-engineering-plugin --skill setup- SKILL.md5.4 KB
Overview
This skill configures which review agents run for your project and writes a compound-engineering.local.md file in the repo root. It auto-detects the project stack, offers an interactive flow to customize focus and review depth, and persists a reproducible agent list and plan settings.
How this skill works
It looks for an existing compound-engineering.local.md and offers to view, reconfigure, or keep it. It auto-detects the project type (Rails, Python, TypeScript, JavaScript, or general) from common files and then either auto-configures sensible defaults or walks you through stack selection, focus areas, and review depth. Finally it composes the agent lists, writes compound-engineering.local.md, and summarizes the saved configuration.
When to use it
- Initial project onboarding to establish automated reviewers
- Whenever you want to change which agents run during /workflows:review or /workflows:work
- After adding a new framework or major language change to the codebase
- When you need a repeatable, team-shared review configuration file
Best practices
- Run the setup from the repository root so detection reads project files correctly
- Use Auto-configure for quick sensible defaults, customize when you need tailored agents
- Keep the Review Context section updated with project-specific instructions and runtime constraints
- Re-run setup after major architecture changes or when adding new priorities (security, performance)
- Choose ‘Thorough’ for critical releases and ‘Fast’ for quicker iterative reviews
Example use cases
- A Rails app: auto-detects rails and enables DHH-style and kieran Rails reviewers plus security and performance agents
- A TypeScript service: selects TypeScript-specific reviewers and code simplicity plus optional performance checks
- A small Python library: pick Python reviewers with focused security and simplicity scans for fast PR reviews
- A general codebase: use code-simplicity, security, performance, and architecture strategists when no specific stack is detected
- Switching to comprehensive audits before a major release to include git-history-analyzer and data-integrity-guardian
FAQ
You can view the current file, re-run the interactive setup to overwrite, or cancel to keep it unchanged.
How are agents chosen for each stack?
Each stack maps to curated stack-specific reviewers. Focus area selections add corresponding agents. Depth determines whether extra comprehensive agents (git-history-analyzer, data-integrity-guardian, agent-native-reviewer) are included.