Hackathon Playbook — Testing skill for Claude Code
Agent skill for winning judged builds (hackathons, game jams, content challenges, take-homes): decode the event, kill-test ideas, build, verify, demo, and defend every decision.
How to install Hackathon Playbook
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open SidTheKid-dotcom/hackathon-playbook and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Hackathon Playbook does
Agent skill for winning judged builds (hackathons, game jams, content challenges, take-homes): decode the event, kill-test ideas, build, verify, demo, and defend every decision. Works with Claude Code, Codex, Cursor.
Alternatives in Testing
- Develop Web Game — Build and test web games iteratively using Playwright with time-stepping 14.6k ★
- Deslop Reborn — One iteration of the IronClaw Reborn de-slop loop — take ONE Reborn crate, fan out parallel review sub-agents 12.6k ★
- Claude Code Spec Workflow — Automated Kiro-style Spec workflow for Claude Code 3.6k ★
README
hackathon-playbook
An agent skill for any build that someone judges: hackathons, game jams, content and creative challenges, and take-home assignments. It takes an AI coding agent from the event page to a submission it can defend.
It gives the agent a direction, a structure and a working philosophy, not a script:
- Win on the criteria. Decode the rules, judges and past winners first. Then every hour moves a judging criterion.
- Adapt to the medium. The agent finds out what entries in this event are made of (an app, a game, 3D, hardware, content, research), how the best work in that craft wins, and which tools it uses.
- Never fake anything. Every number traces to a source. There is no mock data in the demo.
- Show evidence, not "done". Tests that can fail, screenshots, measured numbers, live links.
- Defaults, not cages. Every step has a reason. The agent scales the process to the request and the time left. Only seven hard rules never bend.
Set up with your agent
Copy and paste this into your agent:
Read https://raw.githubusercontent.com/SidTheKid-dotcom/hackathon-playbook/main/INSTALL.md and follow it to install the hackathon-playbook skill for me. Reuse an existing install if there is one, verify it loaded, then ask me for my event link.
Works with Claude Code, Cursor, Codex, OpenCode and other agents with a terminal. Prefer doing it yourself? See Install below.
Install
With the [skills CLI](https://github.com/vercel-labs/skills), for Claude Code, Codex, Cursor and other agents:
npx skills add SidTheKid-dotcom/hackathon-playbook
Or by hand. Clone it into your agent's skills folder:
git clone https://github.com/SidTheKid-dotcom/hackathon-playbook ~/.claude/skills/hackathon-playbook # Claude Code
git clone https://github.com/SidTheKid-dotcom/hackathon-playbook ~/.agents/skills/hackathon-playbook # agents that read ~/.agents/skills
Use
Talk to your agent normally. The skill triggers on
Related Skills
Content Coach
A personal LinkedIn content coach for Claude Code - your playbook, your voice, the 3-2-1 system, the slide tes
Rimworld Mod MCP
Local-only MCP server for RimWorld modding: search the game's Defs and C# API, decompile installed mods, build
Validate Hypothesis
Business hypothesis validation skill. Validates ideas through 6 phases (origin check, market confirmation, int
Cuttle
cuttle is a browser for your agent. Websites do not block it. It keeps your logins. When a website asks for a
AI Agent Phases
A phase-by-phase, spec-driven setup for AI coding agents. Four plain files (AGENTS.md, PHASES.md, PROGRESS.md,
SEO Content Skills
Complete SEO content production suite for AI agents — writing, review, QA, and post-publish verification. Work
Related Agents
Pitch
Pitch deck + demo video editor for hackathon. Use for drafting the 90-second demo script, building the Remotio
Idea Generator
Generates hackathon project ideas from one assigned angle. Spawned in parallel, one per angle, during win-hack
Hackathon Judge
Independent hackathon judge with no stake in the build. Given a judge prompt and evidence pack, it scores the