search-console-readiness
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of markdown-based instructions and contains no executable code, scripts, or binaries.
- [SAFE]: The skill includes strong safety guidelines, explicitly prohibiting the agent from inventing verification tokens, signing into user accounts, or taking actions without explicit permission.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing external data such as production URLs and owner-provided tokens.
- Ingestion points: The agent analyzes production URLs, rendered HTML, and verification tokens provided by the owner as described in SKILL.md.
- Boundary markers: The skill requires the agent to label all findings with evidentiary markers (e.g., OBSERVED, SOURCE-INDICATED, DECLARED) to distinguish between direct observation and external claims.
- Capability inventory: The skill describes preparing verification files or meta tags for the repository but does not include tools for arbitrary shell execution or network exfiltration.
- Sanitization: The agent is explicitly instructed to only use tokens provided by the owner, to verify them against the production environment, and is forbidden from inventing tokens.
Audit Metadata