Turn repeated corrections into automated safeguards, such as code checks, metadata flags, or scripts. This skill helps choose where each lesson belongs and turns recurring mistakes into concrete changes or actionable tasks.
Apply when you catch yourself writing the same instruction a second time, or notice a recurring correction. Encode the rule as a lint, metadata flag, runtime check, or script instead of more text.
Replaces repeated written corrections and reminders with mechanisms that prevent mistakes. Options include lint rules that check code without running it, metadata flags, runtime checks, and automation scripts. When several approaches are possible, it chooses the strongest way to prevent the error in that situation.
Use it when you write the same reminder a second time or repeatedly receive the same feedback from someone else. It reviews corrections and test failures to distinguish isolated events from recurring problems. Lessons are then directed to notes, skills or lint rules, or broader principles, depending on their scope. The work goes beyond recording a lesson: it applies the change immediately or leaves a concrete task to carry it out.
Instructions that require judgment cannot always become automated safeguards. In those cases, make the instruction more prominent and add a failure example. When a mechanism can address the problem, use that mechanism alone and remove duplicate written instructions.
This overview was written by AI (OpenAI Codex) from the public SKILL.md. Check the original below for the exact behavior.
AI rated this skill on four points using only the contents of its SKILL.md. Stars and popularity are not taken into account.
The total counts each point as A = 25, B = 15, C = 5. Rated by AI (OpenAI Codex) as of Oct 21, 2026. Treat it as a guide and read the original before installing.
Review the instructions the agent will receive before you install.
Encode recurring fixes in mechanisms (tools, code, metadata, automation) instead of textual instructions. Every error, human correction, and unexpected outcome is a learning signal. Capture it, route it, and close the loop.
Why: Textual instructions are easy to miss. They require the reader to notice, remember, and comply. Structural mechanisms (lint rules, metadata flags, runtime checks, automation scripts) enforce the rule without cooperation.
Pattern: When you catch yourself writing the same instruction a second time:
Pick the strongest mechanism. When more than one mechanism would work, choose the strongest the situation allows (an unrepresentable state that cannot compile, then a lint or banned API that fails CI, then a canonical helper, then a runtime check), because agents copy whatever the surrounding code already does and a weaker guard becomes the next template.
Corollary: If the fix is structural, only use the structural fix. The instruction is the symptom.
Feedback loop:
Anti-patterns:
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Find a trial Storybook package for a specific branch, or trigger its publication when needed. Check release progress and obtain the package reference or commands for creating or upgrading a project with that build.
Look up a project's existing UI components in Storybook, including their props, or configuration inputs, and usage examples. Use the findings to answer questions and identify components to reuse before building duplicates.
Review, improve, rewrite, or draft Storybook documentation in /docs. Assess each page’s purpose, audience, structure, and examples, then make suitable edits or provide a content plan, with formatting checks after changes.
Check out a GitHub pull request, fix linting and TypeScript errors, and commit and push the changes. The workflow supports forked pull requests and repeats type checks to verify fixes while keeping changes focused on these errors.
Organize GitHub issues and pull requests found during quality assurance testing. Apply release tracking labels, classify bugs by four severity levels, and label multiple findings while keeping documentation issues and feature requests outside the severity scheme.
Work through unresolved GitHub pull request reviews one at a time. The skill explains each comment, proposes a fix, asks how to proceed, and handles chosen edits, replies, commits, and thread resolution.