doc
Generate project README, technical docs, and DoraHacks submission materials
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Generate project README, technical docs, and DoraHacks submission materials
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Search and install agent skills from ClawHub, the public skill registry.
Schedule reminders and recurring tasks.
Remote-control tmux sessions for interactive CLIs by sending keystrokes and scraping pane output.
Instinct-based learning system that observes sessions via hooks, creates atomic instincts with confidence scoring, and evolves them into skills/commands/agents. v2.1 adds project-scoped instincts to prevent cross-project contamination.
Test the generated project inside an E2B cloud sandbox via Claude Code, apply fixes iteratively, and produce a quality report
Research a hackathon using a user-supplied URL plus optional doc URLs (sitemap-driven scrape).
| name | doc |
| description | Generate project README, technical docs, and DoraHacks submission materials |
| metadata | {"openclaw":{"always":false}} |
Generate all materials needed for a compelling hackathon submission: README, technical architecture doc, and DoraHacks BUIDL description.
After the project passes testing (test_results.json status is "pass" or "partially_ready").
[DOC AGENT]
Goal: Generate complete submission documentation for the hackathon project.
TOOL CONSTRAINTS: DO NOT call web_search — it is not configured. Use web_fetch(url) only for specific known URLs. All required context is in local workspace files.
Step 1 — Load all context:
read_file("hackathon/selected_idea.json")
read_file("hackathon/plan.md")
read_file("hackathon/test_results.json")
list_dir("hackathon/project/")
Step 2 — Read key source files to understand what was actually built:
Read the main entry point, key modules, and any README that exists.
Note: document what IS built, not what was planned.
Step 3 — Generate README.md:
Structure:
# {Project Name}
> {One-line tagline}
## The Problem
{Clear problem statement — 2-3 sentences, specific and relatable}
## Our Solution
{How it works — what the agent does, what the user experiences}
## Architecture
{ASCII diagram of key components}
## Sponsor Technology Integrations
### FLock.io (Gold Sponsor)
{Exactly how FLock is used — specific API calls, models, role in system}
### Virtuals Protocol (Bronze Sponsor)
{Exactly how Virtuals is used}
### Unibase (Bronze Sponsor)
{Exactly how Unibase is used}
## Demo
{Step-by-step instructions to reproduce the demo, 5-10 steps}
## Tech Stack
{Clean table: Component | Technology | Why}
## How to Run
```bash
git clone ...
pip install -r requirements.txt
cp .env.example .env # Fill in API keys
python main.py
This project was built autonomously by 0xClaw, an AI agent system built on the 0xClaw agent system (OpenClaw ecosystem). 0xClaw participated in and submitted to this very hackathon.
Built with: FLock.io | Virtual Protocol | Unibase
Step 4 — Generate SUBMISSION.md:
Step 5 — Generate PITCH.md:
Step 6 — Write all files to hackathon/submission/: write_file("hackathon/submission/README.md", ...) write_file("hackathon/submission/SUBMISSION.md", ...) write_file("hackathon/submission/PITCH.md", ...)
Also copy README.md to hackathon/project/README.md
## Output Files
- `hackathon/submission/README.md`
- `hackathon/submission/SUBMISSION.md`
- `hackathon/submission/PITCH.md`
- `hackathon/project/README.md` (copy)