geo
Pass
Audited by Gen Agent Trust Hub on May 1, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses
WebFetchandcurlto retrieve content from external websites and sitemaps for SEO auditing purposes. - [COMMAND_EXECUTION]: The tool utilizes
Bashfor orchestration and executes a local Python script~/.claude/skills/geo/scripts/generate_pdf_report.pyto compile audit results into PDF reports. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from the web.
- Ingestion points: External URLs fetched via
WebFetchorcurlas defined inSKILL.md. - Boundary markers: None explicitly defined in the orchestration logic to isolate fetched content from agent instructions.
- Capability inventory: The skill has access to
Bash,Write,Read, andWebFetchtools. - Sanitization: No specific sanitization or filtering of the fetched HTML content is mentioned before it is processed by subagents.
- [SAFE]: No evidence of malicious obfuscation, hardcoded credentials, or unauthorized persistence mechanisms was detected. The data storage in
~/.geo-prospects/is consistent with the tool's stated CRM-lite functionality.
Audit Metadata