Never Again
Capture a bug or mistake as a permanent guardrail so it cannot happen twice. Use this skill whenever a bug is fixed, a test fails and is repaired, the user says "you did that again", "don't do that", "remember this", "never do X", or any correction is made that a future session would otherwise forget. Also use when the user asks to review, promote, or clean up existing lessons and hooks.
Install
npx skills add malaysherasia-ai/claude-never-again --skill never-againgit clone --depth 1 https://github.com/malaysherasia-ai/claude-never-again.git /tmp/claude-never-again
mkdir -p ~/.claude/skills
cp -r /tmp/claude-never-again/skills/never-again ~/.claude/skills/never-againPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Never Again Turn a mistake into something that cannot happen again. Most memory tools write a note and hope it gets read. A note costs tokens on every single turn and can still be drifted past in a long session. This skill asks a different question first: > **Can this be enforced instead of remembered?** If yes, it becomes a hook — a script that runs automatically and blocks the wrong path. Enforced rules cost nothing per turn beyond their one line. If no, it becomes one short line in `LESSONS.md`. ## The rule that governs this skill **This skill must not cost more than it saves.** Every line added to `LESSONS.md` is paid for on every future turn. Be ruthless. If you would not pay 30 tokens per turn forever to prevent this bug, do not file it. ## Workflow ### 1. Establish the lesson You need three things. Get them from the conversation first; ask only for what is genuinely missing. - **Symptom** — what was observed going wrong - **Cause** — the actual root cause, not the surface - **Rule** — the instruction that prevents it, written as an imperative If the cause is still unknown, stop. A lesson without a cause becomes a superstition, and superstitions never expire. **Spend tokens here, not per turn.** A lesson is filed once and read on every turn forever, so the moment of filing is the cheap place to be thorough and the rule line is the expensive place to be vague. Before you write anything down: read the code path that failed, not just the diff
Details
- Skill name
never-again- Repository
- malaysherasia-ai/claude-never-again
- Path
skills/never-again/SKILL.md- License
- MIT
- Repo updated
- 16 October 2026
- Listed in
- karanb192/awesome-claude-skills
FAQ
What does the Never Again skill do?
Capture a bug or mistake as a permanent guardrail so it cannot happen twice. Use this skill whenever a bug is fixed, a test fails and is repaired, the user says "you did that again", "don't do that", "remember this", "never do X", or any correction is made that a future session would otherwise forget. Also use when the user asks to review, promote, or clean up existing lessons and hooks.
How do I install the Never Again skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add malaysherasia-ai/claude-never-again --skill never-again. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Never Again popular?
It lives in malaysherasia-ai/claude-never-again, which has 1 GitHub stars and 1 forks, and is included in 1 curated skill list.
Related skills
Grill
1mtangoz/grill
Grill a decision, plan or forecast. An outside AI judge, a model from a different company than Claude, argues the strongest case for and against it, names the cheapest test that would settle each challenge, and gives a verdict. Use when the user says grill, stress-test, pressure-test, poke holes in, red-team or challenge something they decided or plan to decide, when the weekly review marks a decision for grilling, or when they say look back or paste a decision record to score.
River Trash Boom Robot Design
5dungnotnull/river-trash-boom-robot-design-agent-skill
Autonomous River Trash-Boom Design & Testing — Production-grade harness for Autonomous River Debris-Boom Robotics & Hydraulics evidence-backed analysis with real-time data aggregation, recognized domain methods, and self-improving knowledge pipeline.
Keyboard Shortcuts
nparashar150/claude-keyboard-shortcuts
Audit a web app for missing keyboard shortcuts and implement them end to end: a command registry, context-scoped bindings, a command palette, visible kbd hints, a ? help sheet, invariant tests, and WCAG 2.1.4 compliance. Use whenever the user mentions keyboard shortcuts, hotkeys, keybindings, a command palette, Cmd+K, power-user features, or making an app feel fast, and also when building or reviewing any list, table, modal, form, or editor UI where shortcuts and keyboard access obviously belong even if the user never says the word shortcut.
Lead Free Heat Resistant Ceramics
5dungnotnull/lead-free-heat-resistant-ceramics-agent-skill
Lead-Free Heat-Resistant Ceramic Glaze & Body Formulation — Production-grade Ceramic Materials & Lead-Free Glaze Engineering analysis harness. Use this skill whenever analyzing ceramic materials, formulating lead-free glazes, designing clay bodies, evaluating thermal expansion fit, testing food-safety compliance, or recommending firing schedules. This skill provides evidence-backed, risk-disclosed analysis with authoritative sourcing.
Prove It Ran
agentkitco/agentkit-dev-skills
Verify that tests, builds and checks actually ran and exercised the changed code, before claiming a task is done. Use before reporting completion, or when asked to prove tests ran or verify a fix.
Idea Autopsy
5auracura/idea-autopsy
Stress-test a business idea — especially one an AI suggested — before anyone builds it. Use when the user says "autopsy this", "is this a business or bullshit", "test this idea", or pastes a startup/side-hustle idea and wants to know if it's real. Returns one of four verdicts: BUSINESS, BUSINESS WRONG OWNER (real idea, wrong constraints — says which), BULLETIN (it's really just content or information), or BULLSHIT, plus the smallest test that would prove it wrong. Built from a four-hour session in which an LLM pitched twelve businesses and eleven died.
Browse the most popular Claude skills, all testing & qa skills, or write your own with the skills generator.