code-checkers

Files (1)

• SKILL.md

SKILL.md (first lines)

# Post-Edit Verification Rules

Whenever you write, edit, or modify code, **you must absolutely run** the linter, formatter, and type checker for that environment to ensure you haven't introduced silent errors or code violations.

*Note:* Do not ask the user to run these commands. Do it proactively using the `run_command` tool (with enough WaitMsBeforeAsync for it to finish) and make sure to deliver code that passes all rules.

## Frontend (TypeScript / React)

When editing code in the frontend (`web/src/**/*.ts`, `*.tsx`), run the following combined command from the frontend root directory (`web/`):

```bash
cd web
pnpm exec eslint src/ --ext .ts,.tsx && pnpm exec tsc --noEmit && pnpm run build
```

### Action after Running (Frontend):
1. **If it passes successfully**: There is no output or it returns successfully. Continue with your workflow.
2. **If there are ESLint errors**: Analyze the output, edit the files to fix the reported issues (e.g., Hooks rules, unused variables, unescaped quotes in JSX, etc.) and **re-run the command**.
3. **If there are TypeScript errors**: Analyze the output, review the interfaces, dependencies, and typings. Modify the code and **re-run the command** until the TypeScript compiler approves the project.
4. **If there are Next.js Build errors**: (e.g. "Event handlers cannot be passed to Client Component props") This means you violated Server/Client component boundaries. Fix the `"use client"` directives and re-run.

## Backend (Python / FastAPI)

When editing code in the backend (`api/src/**/*.py`), run the following combined command from the backend root directory (`api/`):

```bash
cd api
uv run ruff check src/ && uv run ruff format --check src/ && uv run basedpyright src/ && uv run python -c "import src.main"
```

### Optional: Auto-fix
If linting errors are auto-fixable, you may run:
```bash
cd api
uv run ruff check src/ --fix
uv run ruff format src/
```
Then re-run the combined check commands to confirm everything passes.

### Action after Running (Backend):
1. **If it passes successfully**: There is no output or it returns successfully. Continue with your workflow.
2. **If there are Ruff lint errors**: Analyze the output, edit the files to fix the reported issues (e.g., unused imports, missing type hints, naming conventions, etc.) and **re-run the command**.
3. **If there are Ruff format errors**: Run `uv run ruff format src/` to auto-format, then verify with the check command.
4. **If there are Basedpyright errors**: Fix the strict type issues (Pydantic mismatches, `Any` usages, unknown attributes) and **re-run the command**.

From 2004sfm/vaultlock, MIT.

More from 2004sfm/vaultlock