Open-source Agent Skills for Claude Code and Cursor, including implementation-loop, engineering-mode, and web-slides.
olddonkey-skills is easy to set up with trust notes worth reviewing. Check agent compatibility and use-case fit before adding it to your workflow.
npx skills add olddonkey/olddonkey-skillsRun the command in your terminal.
Confirm that the skill files were added to your agent workspace.
Check the README requirements before invoking the skill in your agent.
Repository setup guidance
Looks usable, but maintenance, license, or security notes deserve a closer look.
Last commit was about 24 days ago.
15 GitHub stars indicate community interest.
0 open issues signal maintenance load.
MIT license detected.
3 security/trust notes recorded.
Setup difficulty is 2/5.
A public repository of example skills for Claude, demonstrating how to create reusable instructions and resources for specialized tasks.
Adversarial AI bug hunter with auto-fix skill for Claude Code, Cursor, Codex CLI, GitHub Copilot CLI, Kiro CLI, Opencode, Pi Coding Agent, and more. Multi-agent pipeline finds security vulnerabilities, logic errors, and runtime bugs — then fixes them autonomously on a safe branch.
An open-source framework to turn HTML, CSS, and animations into deterministic MP4 videos.
A multimodal toolset for AI agents to generate, edit, and display images, videos, and audio via CLI.
This repository provides ready-made skills for AI coding assistants. It includes a skill that lets Claude delegate coding to other tools like Codex while keeping review and approval, a skill for goal-first engineering, and a skill to create HTML slide decks. You can install them via a marketplace or manually.
olddonkey-skills is a collection of open-source Agent Skills designed for AI coding assistants like Claude Code and Cursor. The skills are distilled from real-world usage and focus on improving workflow efficiency and code quality. The main skill, implementation-loop, allows Claude to delegate coding tasks to external AI agents (Codex, grok, or cursor-agent) while retaining control over review, testing, and final approval. This separation of concerns ensures that the implementing agent writes code, but Claude handles the judgment and quality gate. engineering-mode is a goal-first engineering skill that emphasizes investigation, design, and planning before implementation. It works in conjunction with implementation-loop to drive unit-by-unit development. For Cursor users, the repository includes a Cursor Plugin that ships both engineering-mode and a plan-first implementation loop, enabling a parent agent to review and publish while a subagent writes code. Additionally, web-slides is a skill that converts material or outlines into interactive 16:9 HTML slide decks for live presentations, featuring 24 built-in themes and a presenter view. Installation is straightforward via the Claude Code marketplace or manual copying. The repository is MIT-licensed and actively maintained, with recent updates and a focus on practical, battle-tested workflows.
15
Stars
1
Forks
0
Issues
MIT
License
Open-source Agent Skills for Claude Code and Cursor, including implementation-loop, engineering-mode, and web-slides.
Developers using Claude Code who want to leverage external AI agents for implementation., Engineering teams seeking structured workflows for AI-assisted development., Technical presenters who need to quickly create slide decks.
The setup section provides repository-level starting guidance, not a guarantee of an independently verified installation. Check the official README and release notes.
Skills execute code and may require API access; review before use. External agents (Codex, grok) may send code to third-party services; ensure compliance.
Similar or complementary options to evaluate include anthropic-skills, bug-hunter, hyperframes.
Aug 20, 2026