skills/verification/SKILL.md
--- name: verification description: Discover and run project validation gates: format, lint, typecheck, LSP diagnostics, tests, build, static security checks, dependency audits, and RTK output handling. Use before claiming work is complete, when fixing broken checks, or when setting up a validation plan. --- # Verification Use this skill to prove changes with the strongest practical checks the repo already supports. ## Discovery Order 1. Read task aliases: `package.json`, `pyproject.toml`, `
npx skillsauth add oornnery/.agents skills/verificationInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Use this skill to prove changes with the strongest practical checks the repo already supports.
package.json, pyproject.toml, justfile, Taskfile.yml, Makefile, CI workflows.task check, pnpm check, make test, or CI-equivalent scripts.Do not install new validation tools just because they are listed here.
Use configured tools only:
uv run ruff format --check .
uv run ruff check .
uv run ty check src
uv run pyright
uv run pytest -v
uv run task sec
uv run bandit -r src
uv run pip-audit
Prefer uv run task check or another repo alias when it maps to the same gates.
Use the repo package manager:
pnpm lint
pnpm typecheck
pnpm test
pnpm build
pnpm exec playwright test
pnpm audit --prod
Use npm, yarn, or bun only when their lockfile owns the repo.
Use LSP diagnostics when the environment exposes them. Treat them as another signal, not a replacement for command-line checks.
Record:
Run static security checks when code touches auth, permissions, secrets, file handling, templates, webhooks, subprocesses, external URLs, SQL, deserialization, dependency changes, or production config.
Security checks may include:
task secskills/security/SKILL.mdDo not hide findings inside generic "lint passed" wording. Triage them separately.
Use RTK for noisy output when available:
rtk <command>
rtk proxy <command>
rtk gain
Use raw commands when full unfiltered output is needed for diagnosis.
Report checks as facts:
Command: ...
Result: PASS | FAIL | SKIPPED
Reason: ...
Risk: ...
Never say validation passed if a required check was skipped, unavailable, or failed.
tools
Build, review, or validate standalone Python scripts run with uv inline metadata. Use for one-file automation, operational scripts, script dependencies, shebangs, idempotency, safety, representative runs, and promoting scripts to packages.
development
Build, review, or validate Python packages and libraries where public API stability, packaging metadata, imports, examples, changelogs, build output, and compatibility matter.
tools
Build, review, or validate Python command-line applications and terminal tools. Use for argparse, Typer, Rich, Textual-adjacent CLI UX, stdout/stderr contracts, exit codes, automation-friendly flags, help output, and CLI tests.
testing
Manage lightweight project state files such as SPEC.md, DESIGN.md, TODO.md, .spec/, and .mem/. Use when starting multi-step work, planning features, recording decisions, updating next steps, preserving cross-session context, or preparing handoff notes.