competitive-visual-audit
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- NO_CODE (SAFE): The skill consists entirely of Markdown text and templates. There are no scripts (.py, .js, .sh) or executable files included.
- INDIRECT_PROMPT_INJECTION (LOW): The templates encourage the agent to ingest external data such as competitor website language and user reviews. While this is a common ingestion surface, the skill lacks any capabilities like network access, file writing, or command execution to exploit, neutralizing the risk. 1. Ingestion points: reference/templates.md (Individual Competitor Profile Template fields for 'Sample Language' and 'User Perception'). 2. Boundary markers: Absent. 3. Capability inventory: None. No code or subprocesses detected in SKILL.md or reference/templates.md. 4. Sanitization: Absent.
Audit Metadata