okf-build
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns such as hardcoded credentials, remote code execution, or persistence mechanisms were detected. The skill's operations are confined to calling specific MCP tools (seo_list_reports, seo_run_report) and processing their outputs.
- [PROMPT_INJECTION]: The skill processes untrusted content from external web crawls, representing an indirect prompt injection surface. 1. Ingestion points: Crawl-derived concepts and page inventory as described in SKILL.md. 2. Boundary markers: None are explicitly defined to separate untrusted data from agent instructions. 3. Capability inventory: The skill uses seo_run_report to generate files and instructs the agent to explain evidence and limits. 4. Sanitization: No specific filtering or escaping of the crawl data is mentioned before processing. This surface is inherent to the skill's primary purpose and is managed by the agent's internal guardrails.
Audit Metadata