crawl
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill makes network requests to
https://api.inspiro.topto perform crawling and mapping operations. - [COMMAND_EXECUTION]: The provided
scripts/crawl.shexecutes several shell utilities includingcurl,jq,sed,mkdir, andbase64. These are used to interact with the API, parse JSON data, and manage local files. - [PROMPT_INJECTION]: The skill identifies as having an indirect prompt injection surface because it ingests untrusted data from external websites that may be subsequently processed by an AI agent.
- Ingestion points: Website content is retrieved from external URLs and processed in
scripts/crawl.sh. - Boundary markers: No specific boundary markers or delimiters are added to the saved markdown files to differentiate between source content and instructions.
- Capability inventory: The skill can perform network requests and write files to the local directory provided by the user.
- Sanitization: The script performs filename sanitization using
sedto prevent path traversal, but it does not sanitize the body of the retrieved web content.
Audit Metadata