foryourhealth111-pixel/Vibe-Skills / bundled/skills/claude-skills
claude-skills
VerifiedClaude Skills meta-skill: extract domain material (docs/APIs/code/specs) into a reusable Skill (SKILL.md + references/scripts/assets), and refactor existing Skills for clarity, activation reliability, and quality gates.
| File | Kind | Size |
|---|---|---|
| AGENTS.md | markdown | 1.8 KB |
| SKILL.md | skill md | 7.9 KB |
| assets/template-complete.md | markdown | 2.4 KB |
| assets/template-minimal.md | markdown | 906 B |
| references/README.md | markdown | 7.5 KB |
| references/anti-patterns.md | markdown | 3.0 KB |
| references/index.md | markdown | 1.2 KB |
| references/quality-checklist.md | markdown | 1.9 KB |
| references/skill-spec.md | markdown | 4.7 KB |
| scripts/create-skill.sh | script | 4.1 KB |
| scripts/validate-skill.sh | script | 5.4 KB |
references/anti-patterns.md
View on GitHub# Anti-Patterns (And How to Fix Them)
This file documents common ways Skills fail in practice. Use it when refactoring existing Skills.
## 1) Documentation Dump in Quick Reference
**Symptom**
- Quick Reference contains paragraphs of text, pasted docs, or large excerpts.
**Why it fails**
- Users cannot scan or reuse it.
- The model treats it as "knowledge soup" rather than executable patterns.
**Fix**
- Move long text into `references/` (split by topic).
- Keep Quick Reference as <= 20 copy/paste patterns when possible.
- Add Examples for anything non-trivial.
## 2) Vague Triggers ("Helps with X")
**Symptom**
- `description` says "helps with databases" or similar.
- "When to Use" is a generic list with no tasks/inputs/goals.
**Why it fails**
- Activation becomes noisy and unpredictable.
**Fix**
- Rewrite `description` as: "What + when".
- In "When to Use", list decidable tasks:
- "Writing migration SQL for PostgreSQL"
- "Debugging a failing CCXT order placement"
- Add "Not For / Boundaries" to prevent misfires.
## 3) Missing Boundaries
**Symptom**
- The Skill never says what it will not do.
**Why it fails**
- The model over-promises and makes unsafe assumptions.
- The skill triggers in irrelevant contexts.
**Fix**
- Add `## Not For / Boundaries` with:
- explicit out-of-scope items
- required inputs and what questions to ask when missing
## 4) Non-reproducible Examples
**Symptom**
- Examples are pseudo-code, missing inputs, or missing expected outputs.
**Why it fails**
- Users cannot trust or validate the behavior.
**Fix**
- Each example should contain:
- Input(s)
- Steps
- Expected output / acceptance criteria
- Prefer minimal reproducible examples over big "showcase" code.
## 5) One Giant File Syndrome
**Symptom**
- Everything is in `SKILL.md` (or one huge reference file).
**Why it fails**
- The entrypoint becomes unscannable and hard to maintain.
**Fix**
- Keep `SKILL.md` execution-focused (patterns + examples + boundaries).
- Split long content into `references/` and add `references/index.md`.
## 6) Invented Facts and Unverifiable Claims
**Symptom**
- The Skill claims API fields/flags/commands without citing sources.
**Why it fails**
- Incorrect guidance is worse than missing guidance.
**Fix**
- Add a "verification path": where/how to confirm in official docs or source code.
- Prefer statements backed by your material; mark assumptions explicitly.
## 7) Unsafe Defaults and Destructive Commands
**Symptom**
- The Skill suggests destructive commands as the default path.
**Why it fails**
- Users copy/paste and lose data.
**Fix**
- Put destructive actions behind explicit warnings and confirmation steps.
- Prefer read-only diagnostics first.
## 8) Inconsistent Terminology
**Symptom**
- Same concept has multiple names; different concepts share a name.
**Why it fails**
- Increases cognitive load and produces inconsistent outputs.
**Fix**
- Add a short glossary (in `references/getting_started.md` or similar).
- Use one concept, one name.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 198k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |