competitive-intelligence
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from untrusted external sources, which is a standard surface for indirect prompt injection.
- Ingestion points: The skill performs web research on competitor websites, review platforms (G2, Capterra, TrustRadius), blogs, and release notes. It also supports optional connectors for CRM data, internal documents, and communication transcripts (Slack, call recordings).
- Boundary markers: The instructions do not explicitly define delimiters or specific "ignore instructions" warnings for the ingested external data.
- Capability inventory: The skill is restricted to research, data aggregation, and the generation of a local HTML artifact. No capabilities for remote code execution, system-level modifications, or unauthorized network operations were identified.
- Sanitization: The skill description does not mention specific sanitization or filtering logic for the data gathered from the web before it is rendered into the final HTML battlecard.
Audit Metadata