blog-outline
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for SERP-informed outline generation and follows standard data processing patterns for this use case.
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection due to its core functionality of processing untrusted web data.
- Ingestion points: Step 2 uses
WebSearchandWebFetchto collect external data from top search results. - Boundary markers: None present in the instructions to separate fetched data from the prompt context.
- Capability inventory: Includes file system access (
Write,Bash,Read,Grep,Glob) and web access tools (WebSearch,WebFetch). - Sanitization: No explicit sanitization of the retrieved web content is performed; the agent is instructed to summarize it into a structured markdown outline.
Audit Metadata