AI Agent Skills
Smello ships with two Agent Skills that work with Claude Code, Cursor, GitHub Copilot, and 20+ other AI coding tools.
Install
Install both skills at once:
Or install them individually:
npx skills add smelloscope/smello --skill smello-setup
npx skills add smelloscope/smello --skill smello-debugger
The command places skills in your project's .claude/skills/ directory (or equivalent for other agents). They're available immediately.
Available skills
/smello-setup: Project integration
Explores your codebase and sets up smello run as part of your development workflow. The skill:
- Detects your package manager (pip, uv, poetry, pipenv)
- Finds HTTP, gRPC, and logging library usage (
requests,httpx,aiohttp,grpc,botocore, Google Cloud libraries,logging) - Locates how you run your app (Django, Flask, FastAPI, CLI scripts, etc.)
- Checks for Docker Compose files and run scripts
- Proposes wiring
smello runinto your dev commands, adding FastAPI middleware if applicable, and enabling log capture
The skill does not make any changes: it only proposes. You approve each step before anything is modified.
Usage (in Claude Code or any compatible agent):
/smello-debugger: Debug with captured events
Helps you debug issues using events captured by a running Smello instance. The skill queries the Smello API to:
- List recent events with filtering (event type, host, method, status code, full-text search)
- Fetch full event details (headers/bodies for HTTP, tracebacks/frames for exceptions, message/extra for logs)
- Diagnose failures: API errors (4xx/5xx), unhandled exceptions, error-level log patterns, slow requests
- Cross-reference captured events with your source code to pinpoint root causes
This skill is also invoked automatically when you ask your AI agent about debugging: whether it's an API error, a crash, a log message, or unexpected behavior.
Usage:
Or just ask naturally:
Why is my Stripe API call returning 401?
Show me the recent exceptions
What errors are in the logs?
What HTTP calls is my app making to OpenAI?
How skills work
Skills follow the Agent Skills open standard. They are Markdown files (SKILL.md) that give your AI agent specialized knowledge and workflows. Unlike plugins or extensions, skills require no runtime dependencies: they are pure instructions.
After installing with npx skills add, the skill files are placed in your project's .claude/skills/ directory (or equivalent for other agents). Your AI agent loads them automatically.
Compatibility
Skills work with any AI coding tool that supports the Agent Skills standard: