rune-verification

Runs lint, type-check, tests, and build checks to verify code health after changes.

1|Updated Mar 22, 2026
One-click install
npx skills add https://github.com/dangvu008/VietTruyen --skill rune-verification-dangvu008
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rune-verification
Source: https://github.com/dangvu008/VietTruyen/tree/main/.agents/skills/rune-verification
Command: npx skills add https://github.com/dangvu008/VietTruyen --skill rune-verification-dangvu008

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After any code change, developers need confidence that nothing broke. This Skill automates the full verification cycle—lint, type-check, tests, and build—across multiple languages, and produces an evidence-backed report instead of unverified claims of success. ## Core Features & Use Cases - Multi-Language Verification: Detects project type (Node.js, Python, Rust, Go, Java) and runs the appropriate lint, type-check, test, and build commands. - 3-Level Artifact Verification: Checks that every created file exists, contains substantive logic (not stubs or truncated code), and is actually wired into the codebase. - Evidence-Before-Claims Gate: Rejects any "done" or "passing" claim that lacks actual command output, preventing false verification reports. - Use Case: After refactoring an authentication module, run this Skill to confirm lint passes, types check, all tests succeed, the build completes, and every new file is genuinely implemented and imported. ## Quick Start Run the verification skill to check lint, types, tests, and build for this project and show me the full report.

Frequently Asked Questions about rune-verification

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

FAQPage Schema
How do I verify code changes didn't break anything?▼

Run all four checks: lint, type-check, tests, and build. This Skill detects your project type from config files like package.json or pyproject.toml, executes the appropriate commands, and reports pass/fail with actual command output as evidence.

How to run lint and tests for multiple languages automatically?▼

Detect the project type first, then map to the right tools: ESLint and Vitest for Node.js, Ruff and Pytest for Python, Clippy and cargo test for Rust, golangci-lint and go test for Go. The Skill handles this detection and command selection automatically.

Does this verification work with Python projects without pyproject.toml?▼

Yes, it falls back to setup.py detection and runs python setup.py check. For scripts-only projects with no build config, the build step is marked SKIP rather than FAIL, while lint and tests still run normally.

Why does exit code 0 not mean my build passed?▼

Many tools exit 0 even when they fail silently. Verification must confirm actual output artifacts: a pass-count line from test runners, a non-empty dist directory from builds, or zero-problem counts from linters.

What happens when a verification tool is not installed?▼

The check is marked SKIP with a note that the tool is not installed, not PASS. PASS requires the tool to actually run and report clean results. If any check is skipped without a valid reason, the overall result is INCOMPLETE.

How do I detect stub or placeholder code in generated files?▼

Scan for stub patterns like functions returning only null, components rendering placeholder divs, or TODO-only bodies. Also check that files are imported by at least one consumer, since existing but unwired files are dead code.