che-qa

Detects project tech stack and runs build, lint, typecheck, and test stages with structured reports.

Updated May 12, 2026
One-click install
npx skills add https://github.com/laionazeredo/che-ai --skill che-qa-laionazeredo
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: che-qa
Source: https://github.com/laionazeredo/che-ai/tree/main/skills/che-qa
Command: npx skills add https://github.com/laionazeredo/che-ai --skill che-qa-laionazeredo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Running quality gates manually across heterogeneous projects (Nx monorepos, Rust, Go, Python) is error-prone and inconsistent. This Skill automates stack detection and executes build, lint, typecheck, and test stages in the correct order, returning a structured numbered report so developers can fix failures without guessing. ## Core Features & Use Cases - Automatic Stack Detection: Scans the worktree for signals like package.json, Cargo.toml, go.mod, nx.json, and biome.json to build a stack profile and pick the right commands. - Ordered Quality Gates: Runs Stage A (build) through Stage E (test naming lint), stopping at the first failure with a formatted report including root cause hypotheses and reproduction commands. - Evidence Retention Outside the Worktree: Writes QA reports, screenshots, and SHA256 evidence manifests to session and workspace directories, never polluting the user's repository. - Use Case: After a developer completes a task in a pnpm/Nx monorepo, invoke this Skill to run affected builds, Biome lint, tsc typecheck, and Vitest tests, then receive a PASS/FAIL report with an auditable evidence manifest. ## Quick Start Ask the agent to run QA on the current task's modified files in the worktree and return a structured pass or fail report.

Frequently Asked Questions about che-qa

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run automated QA checks on changed files in a monorepo?▼

Detect the stack from signals like nx.json or pnpm-workspace.yaml, then scope execution with nx affected or turbo filters targeting only modified projects. Run build, lint, typecheck, and tests in that order, stopping at the first failing stage.

What order should build, lint, typecheck, and tests run in CI?▼

Run build first to catch compile errors, then lint for style issues, then typecheck for type errors, and finally unit and integration tests. Each stage should abort the pipeline on failure so later stages never run against broken code.

Does this QA workflow support Rust, Go, and Python projects?▼

Yes, stack detection covers Cargo.toml for cargo build, clippy, and test; go.mod for go build, vet, and test; and pyproject.toml for compileall, ruff, mypy, and pytest. JavaScript and TypeScript projects use pnpm, Biome or ESLint, tsc, and Vitest or Jest.

Why should QA reports and evidence stay outside the git worktree?▼

Writing reports, screenshots, and logs inside the worktree pollutes the repository and risks committing artifacts. Storing them in session and workspace directories keeps the repo clean while preserving a durable SHA256-hashed audit trail per commit.

When should end-to-end tests be excluded from per-task QA?▼

Playwright or Cypress E2E suites are too slow for per-task gates and should only run when the task itself is E2E-focused. Per-task QA scopes to unit and integration tests affected by the changed files instead.