Progressive-disclosure instructions an agent loads on demand. Cheap until invoked, which is why the per-session figure on each card is usually two digits.
Engineer1999/A-Curated-List-of-Awesome-Claude-Skills
Skill Claude CodeCodex
Automate SEO research with Ahrefs -- analyze backlink profiles, research keywords, track domain metrics history, audit organic rankings, and perform batch URL analysis through the Claude Skill Ahrefs integration.
alexanderop/awesome-claude-code-vue
Skill Claude Code
Use when answering questions about OR implementing features with Nuxt Content (queries, MDC components, document-driven mode), especially when uncertain about current API - Nuxt Content is frequently updated with breaking changes, so fetch current documentation from llms.txt before responding to ensure accuracy over…
machinarii/awesome-mental-models
Skill Claude CodeCodex
Use when the user wants help thinking through a decision, problem, strategy, plan, evaluation, communication, or any open-ended judgment call where a mental model or framework would sharpen the analysis. Triggers include "help me decide", "how should I approach this", "I'm stuck on", "which mental model fits", "apply…
Skill Claude CodeCodex
Use when an AI coding agent must provision, inject, run, or rotate secrets without ever seeing their plaintext — you work with ss:// refs (like ss://stripe/prod/STRIPEWEBHOOKSECRET) while a local daemon resolves the real value at the last possible moment.
Skill Claude CodeCodex
Use when reviewing a code change or diff for correctness, security, missing tests, and convention violations before opening or approving a PR. Review independently and adversarially, then fix high-confidence issues.
Skill Claude CodeCodex
Use when writing Python code - covers typing, error handling, style, imports, and patterns.
Skill Codex
Cross-agent full-stack startup kit for building skill UIs and shippable products. ALWAYS run onboarding FIRST — never scaffold or build until you have interviewed the user (product, scope, style, theme, background animations, architecture, data, auth, payments, launch, deploy) or loaded their stored answers, and…
Skill Claude CodeCodex needs its repo
A required pre-merge quality gate that runs local static checks, tests, and reviews. A merge is allowed only after the checks pass.
Skill Claude CodeCodex
A methodology for designing command-line tools for AI agents. A command-line tool is a program operated through typed commands; these rules focus on discoverable help, structured output, useful errors, safe operations, and reliable retries.
zhangxiaoqiang1991/founders-playbook
Skill Claude CodeCodex
A guided navigation tool for founders building companies around artificial intelligence. It uses a founder-focused playbook to identify the situation and point to a relevant next step.
Skill Claude CodeCodex
Use when a commit is refused with "Protected state edit detected", or before editing AGENTS.md, CLAUDE.md, a lockfile, .forge/config.yaml, .github/workflows, lefthook.yml, .beads/, or a generated harness mirror. Do not use to find a way around the gate — there is no environment variable that authorizes a protected…
Skill Claude CodeCodex
Use when the user wants to "interpret results", "explain the solution", "what does this mean", "translate to business language", "present findings", "sensitivity analysis", "what-if analysis", "how robust is this", "visualize the result", "make a recommendation", or needs to translate a mathematical solution back into…
Skill Codex
Use when a user asks an agent to inspect Flutter app logs, network traffic, SSE/WebSocket streams, current page state, screenshots, or debugging context through flog instead of copying from the TUI.
Skill Claude CodeCodex
Check your work with a verification subagent that reviews diffs, runs builds and tests, and evaluates correctness. Read this file for instructions. Use when asked to "check work", "verify changes", "self-verify", "/check-work", "/check", "/verify", or "/self-verify".
Skill Claude CodeCodex
An English-level assessment assistant that tests university-level English from CEFR B1 to C2, where CEFR is a common scale for language ability.
Skill Claude CodeCodex
Extract text and tables from PDF files, fill PDF forms, and merge multiple PDFs. Use when working with PDF documents or when the user mentions PDFs, forms, or document extraction.
jidechao/Prompt-Best-Practices
Skill Codex
Rewrite a weak or vague prompt into a strong one following the four-part framework (Goal / Context / Output / Boundaries) and six prompting principles. Output is a fixed four-section structure: diagnosis, rewritten prompt, change notes, and assumptions made. Use when the user provides an existing prompt (any language)…
PredictabilityAtScale/promptopskit
Skill Claude CodeCodex
Guidance for creating and editing promptopskit prompt files, defaults, variables, and validation-safe templates.
singlepagestartup/singlepagestartup
Skill Codex
Creates and initializes a new issue in the SPS linear workflow.
Skill Claude CodeCodex
Route Office requests into the correct xlsx, docx, or pptx workflow, while preferring Slidev for new presentation drafts unless .pptx is explicitly required.
Skill Claude CodeCodex
An adapter that connects product-management work to a shared PM Copilot runtime instead of copying its rules into the add-on.
Skill Claude Code needs its repo
Use for governed fixed-option decisions that require named seats, fixed quorum, hard attempt/deadline bounds, durable replay, or explicit human approval/cancel boundaries.
Skill Claude CodeCodex
A Russian-language method for editing, shortening, reviewing, or writing practical non-fiction text. It focuses on usefulness, clarity, logical connection, and clean wording, with separate modes for editing, review, and writing from a brief.
JularDepick/user-thoughts.SKILL
Skill Claude Code
A skill that stores a user's project decisions and constraints in mdbase, a persistent project knowledge store. The records can be reused across sessions and by different coding assistants.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: