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.
Skill Claude CodeCodex
Manage the lifecycle of a project's planning docs (PLAN.md + TASKS.md + an implementer agent). Detects project state and routes to one of five actions — GRILL the user round-by-round to generate the decision record itself (a design-tree interview; the only action that creates material instead of reshaping it)…
Skill Codex
Explicit-invocation-only five-stage workflow for coding tasks. Use this skill only when the user explicitly invokes forge or $forge and wants a request turned into prompt.md, research.md, plan.md, implementation, review.md, and stage commits; do not trigger it from task shape alone.
Skill Claude CodeCodex
Walk through pull request review comments one at a time with the user before making or resolving changes. Uses one orchestrator agent that spawns a fresh subagent per comment with isolated context. Presents click-to-choose polls (AskQuestion) at every decision point so the user picks an approach instead of typing. Two…
Skill Claude CodeCodex
Create structured interactive learning platforms for any field or subject. Use this skill whenever the user wants to build a learning site, educational course collection, study guide platform, or knowledge base about ANY topic — psychology, physics, history, cooking, music theory, programming, medicine, etc. Also use…
Skill Claude CodeCodex
File-based persistent memory for sub-agents and skills — where to store it, memory types (user, feedback, project, reference), save/recall rules, and MEMORY.md indexing. Read this skill when an agent definition says it uses persistent memory, or when saving or recalling learnings across sessions.
valtterimelkko/agent-workflow-skills
Skill Claude CodeCodex
Search YouTube videos by query or within specific channels. Supports date filtering, sort ordering, and duration filters to exclude YouTube Shorts. Returns LLM-optimized JSON with video details.
yellowbee686/everything-analysis-agent
Skill Codex
Use the CBETA MCP server for Chinese Buddhist scripture search, catalog lookup, metadata retrieval, KWIC inspection, and source-text citation. Use when Codex needs live CBETA queries, Buddhist canon text lookup, work/juan/linehead citation, translator or dynasty searches, or CBETA MCP tool guidance.
Skill Claude CodeCodex
A downloader for running activities from COROS Training Hub, the web service that stores COROS training records. It saves activities as FIT files, a standard format for fitness data, and avoids downloading the same activity twice.
Skill Codex
Analyze one selected SkillScope skill-use window by extracting precise constraints from SKILL.md, inspecting Codex or Claude Code trace evidence in stages, writing early UI highlight artifacts, and producing an evidence-backed coverage report. Use when SkillScope launches Codex or Claude Code from the browser after a…
Skill Claude CodeCodex
Security-first dependency hygiene for Node and Python projects. Use when auditing dependencies, verifying automated dependency updates, reviewing or editing manifests and lockfiles, triaging vulnerability findings, handling supply chain incidents, or maintaining long-term dependency safety practices.
Skill Claude CodeCodex
DashClaw platform expert for integration, troubleshooting, and governance. Snapshot-based — prefer live queries via python -m livingcode query, or GET {baseUrl}/api/doctor when Python/livingcode/the repo are unavailable.
Skill Codex
Provides information about how to setup various kinds of projects. Useful for initializing projects based on requirements and scaffolding the project.
mave99a/agent-skill-rpg-game-demo
Skill Claude CodeCodex
An RPG game skill about a character or role named 皇后, meaning “Empress” in Chinese, covering its abilities and current state.
antonio-mello-ai/iterm-ai-agents-bundle
Skill Codex
Control iTerm from an agent: inspect windows/tabs/sessions, focus targets, create tabs/windows/splits/2x2 grids, close scoped targets safely, set iTerm badges, send commands to panes, run health checks, and take screenshots. Use when the user asks to manage iTerm panes, tabs, windows, badges, layouts, or agent…
zasuozz-oss/agentic-superpowers
Skill Claude CodeCodex
Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when…
Skill Claude CodeCodex
End-to-end skill for creating git commits with proper tooling and safety checks. Use when you're asked to create a commit.
Skill Claude CodeCodex
Supervise autonomous execution of a frozen implementation plan (Mode B) - run one slice at a time in a fresh session, enforce the mechanical floor, assess every slice from repository evidence, commission independent review where risk warrants it, and stop for a human when the plan or the floor requires one.
Skill Claude CodeCodex
Write and review documentation for Castor Labs PHP packages that use castor-labs/docs. Use this skill when creating, editing, reviewing, or improving .dev/docs content in a Castor package.
Skill Claude CodeCodex
Agent/skill marketplace discovery for TrustyClaw.
Yet-Another-AI-Project/kiwi-skills
Skill Claude CodeCodex
AI agent development standards using golanggraph for graph-based workflows, langchaingo for LLM calls, tool integration, MCP, and LLM best practices (context compression, prompt caching, attention raising, tool response trimming).
ctresb/hermes-security-auditor
Skill Claude CodeCodex
Use when auditing Rust backend code (Axum/Actix/Rocket/Tower) with PostgreSQL or similar relational stores; async runtimes, SQLx/Diesel/SeaORM, OAuth/OIDC, sessions, multi-tenant/RLS, connection pooling, migrations, crypto/secrets, dependency supply chain, unsafe boundaries, CI/CD, containers/IaC, GraphQL/gRPC…
blackwell-systems/agentskills-cli
Skill Claude Code
Interactive guide for converting skills to progressive disclosure pattern. Use when you have a skill >500 lines and want to split it into core + references for better context economy.
Skill Claude CodeCodex
React Query v4 (TanStack Query) best practices, patterns, and troubleshooting. Use when working with useQuery, useMutation, query invalidation, caching, WebSocket integration, or any async state management in React. Based on TkDodo's comprehensive blog series.
Skill Claude CodeCodex
Run repo work through subagents so the manager stays free for the user. Split work into small tasks, give each worker an isolated git worktree, and require the worker to commit, pass a fresh review, and deliver through the repo's delivery mode. Read project rules from the repo. Trigger: "orchestrate", "fan out…
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: