Prepare release verification from a real change
Create a release checklist tied to changed behavior, dependencies, and rollback evidence.
Browse reusable instructions for your AI tools. Copy and customize a prompt, or download a skill for a supported agent.

Create a release checklist tied to changed behavior, dependencies, and rollback evidence.
Describe event inputs, validation, deduplication, and intended actions before integration work.
Ask OpenClaw to inventory skills and instruction files while protecting private configuration.
Convert a failed tool call into evidence, bounded recovery options, and an honest task status.
Review skill metadata, referenced files, and executable or external-action behavior without running it.
Go from a fuzzy idea to a practical build plan with milestones and clear acceptance criteria.
Review code changes for real bugs, edge cases, and actionable improvements.
Turn an ambitious project into a small, testable build plan.
Design tests around behavior, invariants, and plausible failures.
Diagnose why an OpenClaw skill is missing, overridden, or not being discovered.
Map a repository for Codex before starting a substantial change.
Give Claude Code a focused review workflow for an existing diff.
Investigate a reproducible bug with hypotheses, a minimal fix, and a focused regression check.
Get a focused review of correctness, maintainability, and edge cases, grounded in your code.
Map a product workflow into predictable endpoints, validation rules, and error responses.
Choose meaningful tests around a feature’s actual risks and expected behavior.
Build a useful map of a repository and identify where a specific change belongs.
Investigate failures with evidence, ranked hypotheses, and a minimal regression-tested fix.
Check API contracts for consistent resources, ownership, validation, and failure behavior.
Plan a migration with compatibility checks, verification, and a recovery path.
Check whether a scoped change is ready to ship and what still needs verification.
Review a proposed design for concrete trust boundaries, abuse cases, and missing controls.
Specify a coding task with repository context, acceptance criteria, and a verifiable finish line.
Keep a Cursor edit focused on one behavior with clear boundaries and a regression check.
Try another platform, a broader search, or fewer filters.
Explore all resourcesResource formats: Prompts are ready to copy into a conversation. Skills install into the compatible agent shown on their page.
How it worksInclude the instructions, intended platforms, author, and license. Submissions are reviewed before publication.