What it helps you do
One public HTML page, read-only scrape. No browser interactions, logins, provider tools, forms, or site-wide crawl.
Expected output
- Saved page content
- Original URL and fetch time
- Retrieval limitations
Environment and setup
Upstream documents: Claude Code, Codex, Cursor, Windsurf, OpenCode, OpenClaw, OpenHands, Hermes Agent.
CLI README documents these harnesses. Integration has not been installed or tested here.
- Node.js >=22 for the npm CLI
- firecrawl-cli 1.26.3 (pinned package source)
- For hosted service: Firecrawl account and authentication
- A shell and permission to write local evidence files
Read the pinned original Skill instructions ↗ and setup documentation ↗ before installing. No automatic installation is provided by this site. Review a fixed version and its dependencies before running setup.
Costs and permissions
Credit-metered hosted service; free tier and paid plans exist. Model/agent costs are additional.
Commercial API requirement: Hosted path: yes, the Firecrawl API; paid subscription not necessarily required. Self-hosted custom API endpoint is documented but not evaluated here.
Start with one ordinary HTML URL. Avoid paid query/extraction options, PDFs, broad crawl jobs, and automatic top-ups until the user approves a budget.
- Sends the target URL and options to Firecrawl; retrieved content is processed by the hosted service.
- Uses shell commands and writes local output, conventionally under .firecrawl/.
- The skill permits broad firecrawl CLI commands, including capabilities beyond this starter task. Restrict the task to plain URL scraping.
- Setup can install skills across all detected agents; scope setup to the intended agent. Do not run setup defaults: it can change the agent's web-provider routing.
- Authentication stores credentials and upstream README discloses authentication-time telemetry. Review privacy/settings before signing in.
- Remote provider execution, browser interaction, data terms, and paid calls need separate review.
A bounded starter prompt
Use firecrawl-scrape to read this one public HTML URL: <URL>. Save the result and source URL as local evidence. Use plain scrape only; do not use query mode, crawl, interact, or provider tools. Report missing or partial content. Ask before any paid expansion.
This prompt is an original teaching template. It has not been executed with the third-party Skill.
Limits of the result
- Proving that Google has indexed a page
- Private or authenticated content
- A no-cost guarantee
Source reviewed and Not run describe different evidence. No indexing, ranking, traffic, installation safety or blanket runtime compatibility is established here.
Source and license
Maintainer: Firecrawl. Reviewed version: 1.26.3.
Commit: 21a08f5579e3616dbec0bb28250fab0076a2c1a3
License: ISC. Declared in pinned package.json; LICENSE and LICENSE.md at the repository root returned 404. A standalone license grant for the skill directory was not established. Link-only curation; do not redistribute the skill text.
License source ↗ · Original repository ↗