seo-technical-audit
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFECREDENTIALS_UNSAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill accesses the file path
google-api.jsonto authenticate with Google Services for CrUX and Search Console data. This is a sensitive credential file used for the skill's core functionality. - [COMMAND_EXECUTION]: Spawns subprocesses to run internal Python scripts (
pagespeed_check.py,crux_history.py,gsc_inspect.py) using parameters derived from the user-provided domain. - [PROMPT_INJECTION]: The skill processes untrusted data from the target domain's
/robots.txt,/sitemap.xml, and web pages viaWebFetchandFirecrawltools, creating an indirect prompt injection surface. - Ingestion points: Untrusted data enters the agent context from external websites during the crawl and sense-check phases.
- Boundary markers: There are no explicit instructions to wrap or delimit the external content to prevent the agent from following embedded directives.
- Capability inventory: The agent can execute local scripts and perform network operations using the results of these crawls.
- Sanitization: While the domain name is normalized, there is no mention of sanitizing or escaping the content of the fetched HTML or text files.
Audit Metadata