What it does
Project Bootstrap Skill Pack inspects a local repository and turns it into a structured agent onboarding checklist. It detects Node, Python, Docker, and documentation repo types, recommends setup and validation commands as text output (never executed), checks for missing AGENTS.md guidance, and flags sensitive-looking filenames without reading their contents.
Designed for safe first-pass repo intake: no package managers, shell commands, network calls, or config edits ever run. Output is clean JSON with ok, status, project_types, setup_commands, validation_commands, findings, handoff, and mutated_files: [].
Use cases
- Developers onboarding an agent to an unfamiliar repo who need stack detection, setup hints, and handoff sections before work starts.
- Consultants taking over client codebases who need a quick, evidence-backed repo readiness checklist without executing untrusted code.
- OpenClaw teams standardising agent handoffs who want repeatable AGENTS.md, setup, validation, risk, and rollback prompts for each project.
Example usage
python3 scripts/bootstrap_repo.py /path/to/repo --json
python3 scripts/project_bootstrap.py --repo /path/to/repo --json
python3 scripts/bootstrap_repo.py --self-test --json
Example output
{
"ok": true,
"status": "READY",
"project_types": ["node", "python"],
"setup_commands": ["npm install", "python3 -m venv .venv"],
"validation_commands": ["npm test", "pytest"],
"findings": [],
"mutated_files": []
}
Requirements
- Python 3.10+ recommended; scripts use Python standard library only
- No API keys, model credentials, network access, or third-party packages required
- OpenClaw workspace with local file access to the repository being reviewed
QA and security
This skill has been reviewed and cleared for release through the GetAgentIQ release pipeline with Viper QA and Warlock security sign-off (2026-05-28). Stdlib-only, read-only repo inspection — no command execution, no network calls, no config or cron mutation, mutated_files: [] hardcoded in all outputs.