seo-competitor-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to scrape and analyze external SEO data, creating a potential surface for indirect prompt injection if external websites contain malicious instructions.
- Ingestion points: External content is ingested through search and scraping functions via the
sandbase_runtool, as described inSKILL.mdandreferences/sandbase-api-map.md. - Boundary markers: While the workflow emphasizes "bounded" and "verifiable" outcomes, it lacks specific instructions for using structural delimiters (e.g., XML tags) to isolate untrusted external data from the agent's core instructions.
- Capability inventory: The skill possesses the capability to execute discovered external functions through
sandbase_runbased on ingested data. - Sanitization: There is no mention of sanitizing, filtering, or escaping content retrieved from external sources before it is processed by the agent.
Audit Metadata