Agentic Search Optimizer
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to audit and process content from external websites to evaluate 'WebMCP readiness', which creates an attack surface for indirect prompt injection.
- Ingestion points: The agent is instructed to 'audit actual task flows' and 'Check for WebMCP-related attributes in source HTML', which involves reading potentially untrusted data from any website the user directs it to audit.
- Boundary markers: The instructions do not specify any delimiters or safety warnings to ensure the agent ignores embedded instructions that might be present on the audited pages.
- Capability inventory: The skill produces technical deliverables like code templates, audit reports, and 'Agent Friction Maps' based on the analyzed site data. While it does not request specific shell tools in the frontmatter, the agent's logic for generating these reports is directly influenced by external data.
- Sanitization: There is no mention of sanitizing or validating content retrieved from external sites before using it to generate audit findings or implementation templates.
Audit Metadata