code-change-verification

Runs formatting, linting, type checking, and tests in parallel with fail-fast semantics.

Updated May 23, 2026
One-click install
npx skills add https://github.com/kiranimmadi2/promptforge-ai --skill code-change-verification-kiranimmadi2
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-change-verification
Source: https://github.com/kiranimmadi2/promptforge-ai/tree/main/openai-agents-python/.agents/skills/code-change-verification
Command: npx skills add https://github.com/kiranimmadi2/promptforge-ai --skill code-change-verification-kiranimmadi2

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? It prevents incomplete code changes from being marked done by enforcing the full verification stack—format, lint, typecheck, and tests—before work is considered complete in the OpenAI Agents Python repository. ## Core Features & Use Cases - Ordered Verification Stack: Runs make format first, then executes make lint, make typecheck, and make tests in parallel with fail-fast semantics. - Cross-Platform Scripts: Provides both a Bash script for macOS/Linux and a PowerShell script for Windows with identical behavior. - Heartbeat Progress Updates: Emits periodic status messages while long-running parallel steps are still executing. - Use Case: After modifying runtime code or tests, run the verification script to confirm all checks pass before committing, and get immediate failure logs when any step breaks. ## Quick Start Ask the agent to run the code change verification stack and report any failures from formatting, linting, type checking, or tests.

Frequently Asked Questions about code-change-verification

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

FAQPage Schema
How do I run lint, typecheck, and tests together before committing?

Run the provided script from the repository root: bash .agents/skills/code-change-verification/scripts/run.sh on macOS/Linux, or the PowerShell equivalent on Windows. It runs make format first, then lint, typecheck, and tests in parallel.

How to run make targets in parallel with fail-fast behavior?

The scripts launch make lint, make typecheck, and make tests as separate processes in their own process groups. If any step fails, the remaining steps are terminated immediately and the failing log output is shown.

Does the verification script work on Windows?

Yes, a PowerShell script (run.ps1) provides the same workflow on Windows. Run it with powershell -ExecutionPolicy Bypass -File followed by the script path from the repository root.

When can I skip running the full verification stack?

You can skip it for docs-only changes or repository metadata edits. Any change affecting runtime code, tests, or build/test configuration requires the full stack unless a user explicitly requests otherwise.

Why does the verification script fail before running tests?

The script runs make format first and stops if it fails, so later steps never execute. It also requires make plus perl, python3, python, or uv on Unix to manage parallel process groups.