positioning-messaging
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to ingest and process external data sources that may contain untrusted content.\n
- Ingestion points: Phase 2 (Research) and Phase 3 (Execution) involve processing customer interview transcripts, sales call recordings, and external competitor reviews (e.g., from G2 or Reddit).\n
- Boundary markers: The instructions do not define specific delimiters or security prompts to prevent the agent from following instructions potentially embedded within these external research materials.\n
- Capability inventory: The skill includes a Python script
scripts/check-output.pyfor local file validation and references the use of external APIs for company search and technographics.\n - Sanitization: No explicit sanitization, filtering, or content validation steps are defined for the ingested research data before it is interpolated into the messaging templates.
Audit Metadata