customer-research
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill defines an indirect prompt injection surface by directing the agent to ingest and analyze untrusted data from external sources.
- Ingestion points: Identified in SKILL.md under 'Mode 2: Digital Watering Hole Research' and 'Extraction Framework', as well as the 'references/source-guides.md' file.
- Boundary markers: The instructions lack explicit directives for using delimiters or 'ignore' warnings when processing external data.
- Capability inventory: While the skill is documentation-based, the potential impact of an injection depends on the tools (e.g., shell access, file writes) available to the agent in its execution environment.
- Sanitization: The skill does not provide instructions for validating or sanitizing external data before it is incorporated into the agent's context.
Audit Metadata