Sandbox
@jdrhyne/agent-skills

Reusable skills and prompts for Claude Code and Codex

This repo bundles agent skills, prompts, and platform-specific packages that shape how an AI assistant works on a task. Some folders add tool-backed workflows, some provide pure prompt guidance, and some target OpenClaw or Codex directly.

241 stars29 forksPythonUpdated 18d ago
Who it's for

Builders who want their agent to reuse task-specific skills, prompts, and workflows across projects.

What it delivers

You can add ready-made agent behaviors instead of rewriting the same instructions for every task.

What it does

Universal skill library

Reusable skills in `skills/` for planning, research, design review, reporting, CRM tools, and task coordination.

Pure prompt packs

Instruction-only prompts in `prompts/` for frontend design, humanizing writing, senior engineering judgment, and review.

OpenClaw-specific packages

Specialized folders in `clawdbot/` for OpenClaw workflows like memory, sync, tracking, and release checks.

Codex-specific packages

Codex-focused folders in `codex/` for operating Codex CLI and creating related commands or Gemini handoffs.

Validation and release scripts

Scripts such as `scripts/validate-skills.sh` and `scripts/update-readme.sh` to check skill format and regenerate docs.

How to get it

  1. 1Or clone and add the whole repo
    git clone https://github.com/jdrhyne/agent-skills.git ~/agent-skills
  2. 2Copy skills to your Claude Code instructions directory
    # Copy a specific skill
    cp -r skills/elegant-reports ~/.claude/skills/
    
    # Or symlink
    ln -s $(pwd)/skills/planner ~/.claude/skills/

README

agent-skills

═══════════════════════════════════════════════════════════════════════════════════════
█▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀█
█                                                                                       █
█     ░█████╗░░██████╗░███████╗███╗░░██╗████████╗  ░██████╗██╗░░██╗██╗██╗░░░░░██╗░░░░░  █
█     ██╔══██╗██╔════╝░██╔════╝████╗░██║╚══██╔══╝  ██╔════╝██║░██╔╝██║██║░░░░░██║░░░░░  █
█     ███████║██║░░██╗░█████╗░░██╔██╗██║░░░██║░░░  ╚█████╗░█████═╝░██║██║░░░░░██║░░░░░  █
█     ██╔══██║██║░░╚██╗██╔══╝░░██║╚████║░░░██║░░░  ░╚═══██╗██╔═██╗░██║██║░░░░░██║░░░░░  █
█     ██║░░██║╚██████╔╝███████╗██║░╚███║░░░██║░░░  ██████╔╝██║░╚██╗██║███████╗███████╗  █
█     ╚═╝░░╚═╝░╚═════╝░╚══════╝╚═╝░░╚══╝░░░╚═╝░░░  ╚═════╝░╚═╝░░╚═╝╚═╝╚══════╝╚══════╝  █
█                                                                                       █
█▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄█
═══════════════════════════════════════════════════════════════════════════════════════

                    ╱╲    ★  P O W E R  U P  Y O U R  A G E N T  ★    ╱╲
                   ╱  ╲                                               ╱  ╲
                  ╱    ╲   ┌─────────────────────────────────────┐   ╱    ╲
                 ╱  ▲▲  ╲  │  ◄► JIRA  ◄► GA4  ◄► GOOGLE ADS ◄►  │  ╱  ▲▲  ╲
                ╱  ▲▲▲▲  ╲ │  ◄► GSC  ◄► REMOTION  ◄► PLANNER ◄► │ ╱  ▲▲▲▲  ╲
               ╱   ▲▲▲▲   ╲│  ◄► ORCHESTRATOR  ◄► ZENDESK   ◄►   │╱   ▲▲▲▲   ╲
              ╱            ╲│  ◄► KNOWLEDGE-GRAPH  ◄► NUTRIENT ◄►  │╱            ╲
              ╱════════════╲─────────────────────────────────────╱════════════╲
             ╱  ░░░░░░░░░░  ╲                                    ╱  ░░░░░░░░░░  ╲
            ╱  ░░░░░░░░░░░░  ╲                                  ╱  ░░░░░░░░░░░░  ╲
           ╱▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓╲                                ╱▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓╲
          ╱▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓╲══════════════════════════════╱▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓╲
         ▕▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▏                            ▕▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▏

                         ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓
                         ┃                                       ┃
                         ┃    ★ ★ ★  PRESS START  ★ ★ ★         ┃
                         ┃                                       ┃
                         ┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┛

        ══════════════════════════════════════════════════════════════════════
           OPENCLAW  ●  CLAUDE CODE  ●  CODEX  ●  CURSOR  ●  ANY LLM AGENT
        ══════════════════════════════════════════════════════════════════════

                              ©1987 JDRHYNE ENTERTAINMENT

A collection of AI agent skills and prompts for OpenClaw (formerly Clawdbot), Claude Code, Codex, and other LLM-based coding assistants.

🤝 Contributions Welcome! Have a skill that makes your AI agent smarter? We'd love to include it. See how to contribute or just open a PR.

🦞 Looking for deterministic automation pipelines? Check out lobster-workflows — ready-to-use workflow templates with approval gates, state tracking, and repeatable pipelines for OpenClaw.


Every skill in this repo is scanned by AgentVerus — an open-source security scanner that detects prompt injection, data exfiltration, and hidden threats in AI agent skill files. Live results are published from the latest scan run. View full results →

Nutrient MCP Servers (Builder Resources)

If you're using these skills for document AI workflows, also check the official MCP servers:

Structure

agent-skills/
├── skills/      # Skills with tooling, scripts, or structured workflows
├── prompts/     # Pure instruction prompts (guidance text only)
├── clawdbot/    # OpenClaw-specific skills
└── codex/       # Codex CLI-specific skills

Compatibility

79% of skills work across all platforms (OpenClaw, Claude Code, Codex).

CategoryCountPlatforms
Universal8✅ All
Portable18✅ All (uses read/write/exec)
OpenClaw-only7OpenClaw only

See COMPATIBILITY.md for the full matrix and installation instructions per platform.

ClawHub release provenance

Maintained ClawHub releases use a validation-first, exact-SHA workflow with an independently configured production approval gate. See the ClawHub release guide for the ten-package manifest, offline validation, preview, publication, credential, and deprecation boundaries.

Skills

Skills include tooling, templates, scripts, or structured workflows:

SkillTrust           Description
autoresearch-loopDomain-agnostic metric-driven improvement loop, generalizing Karpathy's autoresearch.
context-recoveryRecover missing conversation context after explicit compaction or truncation, or when the user explicitly asks to recover prior work.
design-auditPremium UI/UX design auditor with Jobs/Ive philosophy.
elegant-reportsGenerate beautifully designed PDF reports with a Nordic/Scandinavian aesthetic.
ga4Read Google Analytics 4 reporting data through the GA4 Data API.
gongGong API for searching calls, transcripts, and conversation intelligence.
google-adsQuery, audit, and optimize Google Ads campaigns.
gscRead Google Search Console properties, Search Analytics, URL Inspection, and sitemaps.
jiraRead, search, draft, create, or update Jira work.
last30daysResearch any topic from the last 30 days on Reddit + X + Web, synthesize findings, and write copy-paste-ready prompts.
nudocsUpload, edit, and export documents via Nudocs.ai.
nutrient-openclawUse the pinned Nutrient OpenClaw plugin to convert, OCR, extract, redact, watermark, sign, or inspect the last-known local credit record ...
parallel-taskCoordinate plan files by launching multiple parallel subagents for unblocked tasks.
plannerCreate structured plans for multi-task projects that can be used by the task-orchestrator skill.
remotionBest practices for Remotion video creation in React — compositions, sequences, animation, timing, and rendering.
salesforceQuery and manage Salesforce CRM data via the Salesforce CLI (sf).
task-orchestratorAutonomous multi-agent task orchestration with dependency analysis, parallel tmux/Codex execution, and self-healing heartbeat monitoring.
zendeskManage Zendesk tickets, users, and support workflows through the Zendesk API.

Nutrient MCP & Document Processing

Prompts

Pure instruction prompts — guidance text that works with any AI agent:

PromptTrust           Description
frontend-designExpert frontend design guidelines for creating beautiful, modern UIs.
humanizerRemove signs of AI-generated writing from text.
munger-observerReview a specific decision, plan, or artifact using bounded evidence, counterevidence, alternatives, opportunity cost, and verifiable nex...
senior-engineeringEngineering principles for building software like a senior engineer.
web-design-guidelinesReview UI code for Web Interface Guidelines compliance.

OpenClaw-Specific (formerly Clawdbot)

Skills designed specifically for OpenClaw:

SkillTrust           Description
auto-updaterAutomatically update OpenClaw and selected skills once daily.
clawdbot-release-checkCheck for new OpenClaw releases and notify once per new version.
clawddocsOpenClaw documentation expert with decision tree navigation, search scripts, doc fetching, version tracking, and config snippets for all ...
gallery-scraperBulk download images from login-protected gallery websites using an attached browser session.
knowledge-graphThree-Layer Memory System — automatic fact extraction, entity-based knowledge graph, and weekly synthesis.
self-improving-agentCaptures learnings, errors, and corrections to enable continuous improvement.
skill-syncSync skills between local installation and the GitHub source-of-truth repository.
todo-trackerPersistent TODO scratch pad for tracking tasks across sessions.

Codex-Specific

Skills designed for Codex CLI (OpenAI's coding agent):

SkillTrust           Description
codexUse when the user asks to run Codex CLI (codex exec, codex resume) or references OpenAI Codex for code analysis, refactoring, or automate...
command-creatorThis skill should be used when creating a Claude Code slash command.
geminiUse when the user asks to run Gemini CLI for code review, plan review, or big context (>200k) processing.

Note: Most skills in /skills/ and /prompts/ are universal and work with Codex too. The /codex/ folder contains skills specifically about operating Codex CLI.


Installation

With OpenClaw

Add skill paths to your OpenClaw config:

skills:
  paths:
    - /path/to/agent-skills/skills/elegant-reports
    - /path/to/agent-skills/prompts/frontend-design
    - /path/to/agent-skills/clawdbot/todo-tracker

Or clone and add the whole repo:

git clone https://github.com/jdrhyne/agent-skills.git ~/agent-skills

With Claude Code

Copy skills to your Claude Code instructions directory:

# Copy a specific skill
cp -r skills/elegant-reports ~/.claude/skills/

# Or symlink
ln -s $(pwd)/skills/planner ~/.claude/skills/

With Codex CLI

Copy skills to your Codex instructions directory:

# Copy Codex-specific skills
cp -r codex/codex ~/.codex/skills/
cp -r codex/gemini ~/.codex/skills/
cp -r codex/command-creator ~/.codex/skills/

# Copy universal skills you want
cp -r skills/jira ~/.codex/skills/
cp -r prompts/humanizer ~/.codex/skills/

# Or symlink the whole repo for easy updates
ln -s $(pwd) ~/.codex/agent-skills

Add to your Codex config (~/.codex/config.toml):

[skills]
paths = [
  "~/.codex/skills/codex",
  "~/.codex/skills/gemini",
  "~/.codex/skills/jira"
]

Manual

Each skill/prompt is self-contained. Copy the SKILL.md file (and any supporting files) to your agent's context.


Skill vs Prompt

Skills (/skills/) include:

  • Scripts or tooling
  • Templates or assets
  • Structured multi-step workflows
  • External API integrations

Prompts (/prompts/) are:

  • Pure instruction text
  • Design guidelines or principles
  • Mental models or frameworks
  • No external dependencies

Creating New Skills

Skill Structure

skills/my-skill/
├── SKILL.md          # Required - main instructions
├── scripts/          # Optional - helper scripts
├── templates/        # Optional - templates/assets
└── README.md         # Optional - additional docs

SKILL.md Format

---
name: my-skill
description: Brief description of what this skill does
---

# My Skill

Detailed instructions for the AI agent...

Prompt Structure

Prompts are simpler — just a single SKILL.md file:

prompts/my-prompt/
└── SKILL.md

Scripts

  • scripts/update-readme.sh — Auto-generate README from skill metadata
  • scripts/validate-skills.sh — Validate SKILL.md format across all skills

🤝 Contributing

Contributions are welcome! Whether it's a new skill, an improvement, or a bug fix.

Quick Start

  1. Fork the repo
  2. Create your skill in skills/, prompts/, clawdbot/, or codex/
  3. Test it with your agent
  4. Submit a PR

Quality Bar

Good skills share expert knowledge — the kind that takes years to learn:

  • ✅ Decision trees for non-obvious choices
  • ✅ Trade-offs only experience teaches
  • ✅ "NEVER do X because..." warnings
  • ❌ Don't explain basics the model already knows
  • ❌ Don't write tutorials for standard libraries

See CONTRIBUTING.md for full guidelines.

Ideas Welcome

Not sure what to build? Some gaps we'd love filled:

  • Integrations (Slack, Linear, Notion API)
  • Framework patterns (Next.js, Remix, SvelteKit)
  • DevOps workflows (Terraform, Kubernetes)
  • Domain expertise (legal, medical, finance)

Open an issue to discuss before building something big!


🦞 Lobster Workflows

Looking for deterministic, repeatable automation pipelines? Check out lobster-workflows — a companion repo of ready-to-use Lobster workflow templates for OpenClaw.

Lobster replaces freeform LLM orchestration with typed pipelines, approval gates, and state tracking. The repo includes workflows for content publishing, PR monitoring, release announcements, multi-agent coordination, and knowledge extraction.


License

MIT — Use these skills however you like.

Author

Jonathan Rhyne (@jdrhyne)

Files in the repo

Repository payload14 top-level entries
  • .github
  • clawdbot
  • codex
  • docs
  • prompts
  • scripts
  • skills
  • .gitignore
  • AGENTS.md
  • CHANGELOG.md
  • COMPATIBILITY.md
  • CONTRIBUTING.md
  • LICENSE
  • README.md

Discussion (0)

Ask about usage, or say what you built with it

Sign in to join the discussion.

No comments yet. Be the first to say what this is good for.

More collections

The awesome collection of OpenClaw skills. 5,400+ skills filtered and categorized from the official OpenClaw Skills Registry.🦞

52k
phuryn/
pm-skills
phuryn/pm-skillsCollections

PM Skills Marketplace: 100+ agentic skills, commands, and plugins — from discovery to strategy, execution, launch, and growth.

26k

Extracted system prompts from Anthropic - Claude Fable 5.1, Opus 5, Claude Design, Claude Code. OpenAI - ChatGPT GPT-6-Astra, Codex. Google - Gemini 3.8 Flash, 3.1 Pro, Antigravity. xAI - Grok, Grok Bot, Cursor, Kimi and more! Updated regularly.

65k
1 add