davidondrej/skills

access to david ondrej's personal agent skills

davidondrej‑skills – Re‑usable “Agent Skills” for AI‑powered assistants

What it is

  • A collection of ready‑to‑use skill packages that describe concrete workflows for AI agents (Claude, Codex, Pi, etc.).
  • Each skill lives in its own folder under skills/ and contains a SKILL.md that tells an agent when to invoke the skill and how to carry it out.

Why it matters

  • Modern AI agents often need a library of repeatable procedures – e.g., “pull the latest docs from a repo”, “summarise a YouTube video”, or “orchestrate a chain of coding agents”.
  • By codifying these procedures as plain‑text skill definitions, the same logic can be shared across projects, tools, or the Cloudroom platform without rewriting prompts each time.

Top‑level structure

Category Typical tasks
agent‑orchestration Run, schedule, delegate to, and benchmark coding agents; build agent‑to‑agent loops.
skill‑authoring Create, improve, publish new skills and their context files.
research‑and‑web Fetch data from the web, APIs, browsers, YouTube (most use DeepAPI).
thinking‑and‑docs Structured thinking, interview scripting, teaching outlines, documentation generation.
ops‑and‑setup Machine, server, security, and tool configuration tasks.

How to use

  1. Clone the repo.
  2. Pick a skill folder that matches your need.
  3. Read the SKILL.md – it contains the exact prompt template and any required parameters.
  4. Feed that prompt to an AI coding agent (e.g., via Cloudroom’s free, open‑source runner) when the task arises.
  5. The agent follows the instructions, returns the result, and you can chain multiple skills together for larger workflows.

Who it’s for

  • Developers building AI‑augmented development tools.
  • Researchers who need reproducible “agentic” pipelines.
  • Anyone using Cloudroom or similar platforms to run Claude, Codex, or Pi agents and wanting a library of vetted, shareable workflows.

Key take‑away davidondrej‑skills is not a model or a library of code‑execution functions; it is a catalog of prompt‑driven, reusable procedures that let AI agents perform common, repeatable tasks reliably across projects.

Related

  • Project
    MengTo/Skills

    A curated library of procedural playbooks and prompts that turn complex design and coding workflows into reusable skills for AI agents to ensure consistent, high-quality output.

  • Project
    kostja94/marketing-skills

    An open-source library of reusable marketing and SEO workflows for AI agents, providing structured methods and verification steps to ensure professional-grade outputs.

  • Project
    chujianyun/skills

    A Chinese‑language repo that bundles dozens of reusable “Agent Skills” – small, documented plugins for Claude Code and similar LLM‑agent platforms. Skills cover wiki look‑ups, prompt optimisation, career assessments, diagram generation, document interpretation, audio‑to‑text, image watermarking, web‑scraping, and service run‑books. They are organized under `skills/`, registered in `config/skills.json`, and installed via Claude Code’s marketplace, an installer agent, or the `openskills` CLI.

  • Project
    cosmicstack-labs/mercury-agent-skills

    An open-source library of standardized SKILL.md playbooks that provide AI agents with curated, production-ready workflows across 23 different categories.

  • Project
    kangarooking/kangarooking-skills

    A collection of structured, open-source Agent Skills compatible with the Agent Skills standard, providing reusable workflows for content creation, image generation, and AI project engineering.