The Agent Skills Directory

[SAFE]: The skill analysis did not reveal any malicious code, obfuscation, or unauthorized privilege escalation.
[COMMAND_EXECUTION]: The skill utilizes the firecrawl command-line tool to perform web scraping and markdown conversion, which is its stated purpose.
[DATA_EXFILTRATION]: There is no evidence of unauthorized data transfer. The tool retrieves information from URLs provided by the user and saves it to a local directory.
[PROMPT_INJECTION]: The skill ingests untrusted content from external URLs (Ingestion point: SKILL.md command firecrawl scrape), creating a surface for indirect prompt injection. However, it lacks instructions to bypass safety guidelines. Sanitization is mentioned as providing 'clean' markdown, though explicit boundary markers are not demonstrated in the usage examples. Capabilities include file system writes via the -o flag.

firecrawl-scrape