# AGENTS.md instructions for Firecrawl CLI for web scraping and markdown extraction

Use Firecrawl CLI for web scraping and markdown extraction only for: Extract web pages into markdown or structured data for agent research and retrieval workflows.

## Default behavior

- Prefer non-interactive commands and stable output formats: markdown, json.
- Start with read-only inspection commands before changing state.
- Show the exact command before destructive or deployment actions.
- Require explicit user approval for commands marked `requires-review` or `destructive`.

## Useful commands

- requires-review: `npx firecrawl crawl https://example.com` — Fetches web content for downstream analysis.
- safe: `npx firecrawl scrape https://example.com` — Extracts a single page before broader crawling.

## Safety notes

- Respect robots, rate limits, and site terms before crawling.
- Agents should crawl a small target set first and store source URLs with extracted content.