serp-analysis
Pass
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs SEO analysis on search engine results. It implements security best practices by including explicit instructions to treat external web content as untrusted evidence, mitigating risks from indirect prompt injection.
- [DATA_EXFILTRATION]: Uses the WebFetch tool to retrieve search engine results pages. This is the primary function of the skill and does not involve accessing or transmitting sensitive local credentials, environment variables, or system files.
- [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: External websites retrieved via the WebFetch tool.
- Boundary markers: Includes a dedicated 'Security boundary' section that explicitly instructs the agent to treat fetched content as evidence only and to ignore any embedded owner overrides or prompt-like directives.
- Capability inventory: Reads from and writes to the local memory/ directory (e.g., memory/research/) to persist analysis results.
- Sanitization: Relies on specific instructions to disregard and flag instructions found within processed web data.
Audit Metadata