website-brief
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates by ingesting untrusted data from external sources, creating a surface for indirect prompt injection.
- Ingestion points: The workflow involves using
agent-browserto fetch content from home pages, service descriptions, and customer review sites, as well as reading files from the local repository likeREADME.mdandPRODUCT.md(SKILL.md, Workflow Step 1). - Boundary markers: The skill employs a structured tagging system (
confirmed,decided,inferred,conflict,missing) which serves to demarcate the source and reliability of information in the generated brief (SKILL.md, Workflow Step 2). - Capability inventory: The skill possesses file-write capabilities for the
.agents/business-context.mdfile and network-fetch capabilities via platform tools (SKILL.md, Workflow Step 7). - Sanitization: No explicit validation or filtering is performed on the content retrieved from external sources before it is analyzed by the agent to populate fields.
Audit Metadata