crawl
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill connects to the
api.inspiro.topdomain to retrieve web content as specified by the user's input.\n- [COMMAND_EXECUTION]: The scriptscripts/crawl.shexecutes common system binaries includingcurlfor API communication andjqfor parsing JSON responses. It also performs filesystem operations such asmkdirandechoto save the results.\n- [DATA_EXFILTRATION]: No evidence of sensitive data theft was found. The script utilizes theINSPIRO_API_KEYexclusively to authenticate with the intended service for the skill's primary function.\n- [PROMPT_INJECTION]: The skill ingests untrusted content from external websites which could serve as a vector for indirect prompt injection.\n - Ingestion points: Website data is retrieved via the
crawlendpoint from user-supplied URLs inscripts/crawl.sh.\n - Boundary markers: None. Content from external pages is written directly to local markdown files without specialized delimiters or instructions for the AI agent.\n
- Capability inventory: The skill has the capability to create directories and write files to the local filesystem.\n
- Sanitization: The script performs basic sanitization of filenames using
sedandcutto remove protocols and replace special characters with underscores, mitigating path traversal attacks.
Audit Metadata