checkpoint

Commits work, updates TODO.md, and verifies lint and tests after each implementation wave.

Updated Feb 26, 2026
One-click install
npx skills add https://github.com/Ouaish-Labs/ouaish.com --skill checkpoint-ouaish-labs
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: checkpoint
Source: https://github.com/Ouaish-Labs/ouaish.com/tree/main/.claude/skills/checkpoint
Command: npx skills add https://github.com/Ouaish-Labs/ouaish.com --skill checkpoint-ouaish-labs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? During multi-wave implementations, progress can be lost or untracked without a consistent way to save work, document remaining tasks, and verify code quality before moving on. ## Core Features & Use Cases - Safe Git Commits: Stages only relevant files (never git add -A), excludes secrets and build artifacts, and commits with a descriptive checkpoint message. - Progress Tracking: Creates or updates a TODO.md file with completed tasks, remaining work, and blockers. - Quality Verification: Runs Ruff lint and pytest inside Docker for Python backends, plus pnpm lint for TypeScript frontends, then reports any new errors. - Use Case: After finishing one wave of a multi-step feature implementation, run this Skill to commit the changes, update the task list, confirm tests pass, and get a summary of what remains. ## Quick Start Ask the AI to create a checkpoint of the current work, update the TODO list, and verify that lint and tests pass.

Frequently Asked Questions about checkpoint

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

FAQPage Schema
How do I save progress during a multi-step implementation?▼

Create a checkpoint by staging only the relevant changed files, committing with a descriptive message, and updating a TODO.md file that lists completed and remaining tasks. This preserves a clean history and clear record of what comes next.

How do I run Python lint and tests inside Docker?▼

Use docker compose exec to run commands inside the running container, such as ruff check and pytest against your app and shared packages directories. This ensures the checks run against the same environment as production.

Why should I avoid git add -A when committing work?▼

git add -A stages everything, including .env files, credentials, node_modules, and unrelated changes from other tasks. Staging specific files keeps commits focused and prevents accidentally leaking secrets or mixing unrelated work.

What should a TODO.md progress file contain?▼

A TODO.md should list completed tasks as checked items, remaining tasks as unchecked items, and a section for blockers or open questions discovered during the current wave. This gives anyone resuming the work immediate context.

When should I create a checkpoint instead of a regular commit?▼

Create a checkpoint after completing a wave in a multi-wave implementation or whenever you need to save progress and document remaining work. Regular commits suit isolated changes, while checkpoints combine committing with progress tracking and verification.