Peak Resource README Generator
Peak Resource README Generator is a local-first Node.js CLI that reads a FiveM resource and drafts the documentation server owners need before installation or release.
What it documents
- Installation order and detected dependencies
- Likely configuration files and settings that need verification
- ACE, group, job, and permission references
- Commands, network events, handlers, and exports
- Troubleshooting guidance and a release checklist
The generator skips .git, node_modules, build output, caches, and symlinks. It does not upload resource source code during the scan. An optional --ai polishing pass supports compatible hosted endpoints, local runtimes, and open-weight model servers.
Quick start
Requires Node.js 20+.
bashnode bin/peak-readme.js --resource "./resources/[peak]/my-resource" --out ./README.generated.md
Use --include-source to print detected metadata as JSON for a review pipeline. Add --ai for optional AI editing assistance.
Provider setup
Hosted providers use PEAK_README_AI_KEY, PEAK_README_AI_ENDPOINT, and PEAK_README_AI_MODEL. Any service exposing the compatible /v1/chat/completions contract works, including managed gateways and self-hosted runtimes.
For local Ollama, start ollama serve, pull a model, and set:
powershell$env:PEAK_README_AI_PROVIDER = "ollama" $env:PEAK_README_AI_ENDPOINT = "http://127.0.0.1:11434/v1/chat/completions" $env:PEAK_README_AI_MODEL = "llama3.2"
The output remains a draft. Never scan or send secrets, private keys, webhooks, or credentials.
Important review step
The output is a documentation draft, not a security audit. Confirm framework compatibility, defaults, load order, permissions, server-side validation, and all undocumented runtime behavior before publishing it.