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
An autonomous supervisor for a coding task that repeatedly drives the work through planning, implementation, review, and completion. It runs within a separate Git worktree, which is an isolated working copy of a repository.
jeromekneip/ai-team-starterkit
Skill Claude Code
Document a repeatable process as a new skill or policy. Use when the owner says "make this repeatable", "document this process", "we should have a procedure for this", or when the same multi-step task has been done twice and will recur.
Skill Claude CodeCodex
Run runtz DevSecOps security scans (SCA, SAST, host packages, container images, Kubernetes) and read runtz documentation. Use when the user wants to find vulnerable dependencies or packages, scan source code for secrets/weak crypto, audit a container image or Linux host for CVEs, check Kubernetes posture, or asks how…
simonstone-dev/simonstone-codex-guidelines
Skill Claude CodeCodex
Use when writing, modifying, reviewing, refactoring, or debugging code with Codex to reduce wrong assumptions, overengineering, unrelated diffs, and unverifiable work by applying Simon Stone's coding-agent guidelines.
Skill Claude CodeCodex
Configure the gleanwork agent baseline in a library or SDK repo — scaffold AGENTS.md, a CLAUDE.md pointer, and skills/, author the library's SKILL.md, and verify it in CI. Use when onboarding a gleanwork OSS repo to the agent setup, adding AGENTS.md or a skill to a repo, or writing/updating the SKILL.md that teaches a…
Skill Claude CodeCodex
Post-task self-evolution analysis — logs operational errors, philosophical patterns, and generates optimization plans for continuous improvement.
Skill Claude Code needs its repo
This skill migrates existing Markdown documentation into a modular Single Source of Truth (SSOT) structure managed by docs-ssot, then builds and validates the output.
remix2111/hermes-agent-advanced-skills
Skill Claude CodeCodex
A skill that automatically adds directory-specific rules to a sub-agent before it works in that directory. These rules can be stored in files such as AGENTS.md or .hermes-rules.
Skill Claude Code
Commit changes across the LiteStack meta-repo and its submodules. Use this when the user asks to commit, save changes to git, or "commit this" while working at the LiteStack root. It commits inside each changed submodule first, then records the updated submodule pointers in the meta-repo, and pushes according to the…
shahid0/mobile-development-ai-skills
Skill Codex
Apply this skill by default for all Dart or Flutter development. Do not merely produce compiling code; produce code with correct boundaries between Dart work and Flutter rendering.
Skill Codex
Initialize a cloned template repository for a concrete project. Use when the repository still contains template placeholders, the project purpose or stack is unclear, the user asks to fill AGENTS.md/README/BOARD/docs, or a new project is being started from this template.
Skill Claude CodeCodex
Develop, run, and verify the KimiHarness Tauri app in this repo — dev loop, build, protocol probing, and the pitfalls already discovered. Use for any work on the KimiHarness codebase.
Skill Claude CodeCodex
Generate a guided onboarding tour of any codebase — architecture map, traced key flows, conventions cheat sheet, ranked reading path, and suggested first tasks. Use this skill whenever the user asks to explain a codebase, onboard onto a repo, document an unfamiliar project, produce an architecture overview, create…
Skill Claude CodeCodex
Design and edit UI layouts in Pencil.app .pen files. Use when working with .pen design files, creating UI mockups, generating code from designs, or when user mentions Pencil app. Supports reading designs, creating/editing layouts, taking screenshots for validation, and extracting design tokens.
Skill Claude CodeCodex
Universal guide for creating well-structured Agent Skills for any technology or domain. Use when creating, writing, authoring, or designing a new Skill. Triggers on requests involving SKILL.md files, skill structure, frontmatter, bundled resources (scripts/references/assets), skill validation, skill packaging, or when…
Skill Claude CodeCodex
Skill "skillcraft" from cloudroad-io/skillcraft, covering skillcraft, install, or: pip install skillcraft, commands and how it works.
Skill Codex
Get parallel second opinions from multiple AI coding agents. Use when the user wants independent reviews, architecture feedback, or a sanity check from other AI models.
cobusgreyling/grok-build-showcase
Skill Claude Code
Calculate and explain a "Grok Readiness Score" (0-100) for a project based on concrete signals: quality of AGENTS.md / project rules, quantity and validity of skills, test/CI presence, git hygiene, ignore configuration, and verification culture. Use when asked to "score", "readiness score", "grok score", "how agent…
Skill Claude Code
Turn any folder (code, docs, papers, images) into a navigable knowledge graph with community detection, an audit trail, and HTML + JSON + report outputs. Run with /graphify.
Skill Codex
Use before repository-specific coding, review, debugging, refactoring, explanation, or planning to resolve agent guidance from local AGENTS.md/CLAUDE.md or shared /.icodebyhand notes. Also use when the user asks to remember, store, organize, update, bootstrap, or revise repo-specific guidance without a clear…
Skill Claude Code
Use after implementation changes to update OKF from git diff or changed files. Extract durable behavior, architecture, domain, data, and playbook changes into relevant OKF documents and log entries. Do not use for temporary task notes or changes with no durable project knowledge impact.
calcrowcrab/agentic-engineering
Skill Claude CodeCodex
Design, build, and operate AI agents end to end: understand the system, deep-research the tasks, then write the instructions from that understanding. Use when writing or fixing a system prompt, CLAUDE.md, or tool descriptions; deciding one agent vs several; debugging an agent that follows instructions poorly…
Indium-AI-Labs/indium-agentkit
Skill Claude CodeCodex
Design immutable event logs, CQRS command/query models, and materialized view projections.
Skill Codex
A review process for deciding whether a project skill still matches the work it is meant to guide. It checks the skill’s trigger text, displayed description, instructions, supporting files, and boundaries with project documentation.
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: