seo-project-setup
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from external sources such as Google Search Console CSV exports and third-party website content (competitors, forums, reviews).
- Ingestion points: External website content fetched via URL (Step 4) and local CSV files stored in the
gsc/directory (Step 6). - Boundary markers: The instructions do not define specific delimiters or instruct the agent to ignore potentially malicious instructions embedded within the processed data.
- Capability inventory: The skill has the capability to create and write to local files (creating a
README.mdand workspace structure) and interact with the HarborRank MCP toolset (Step 5). - Sanitization: There are no explicit instructions to sanitize, escape, or validate the content retrieved from external URLs or CSV files before it is processed by the agent.
Audit Metadata