free-tool-strategy
Pass
Audited by Gen Agent Trust Hub on Mar 7, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists entirely of Markdown and JSON files. It does not include any scripts, shell commands, or binary executables, which significantly reduces its attack surface.
- [PROMPT_INJECTION]: Analysis of the instructions shows a standard role-play prompt for an expert strategist. There are no attempts to bypass safety filters or extract system prompts. The skill references reading a context file (.agents/product-marketing-context.md) for business context, which is a standard pattern and poses minimal risk as the skill lacks execution capabilities. 1. Ingestion points: .agents/product-marketing-context.md. 2. Boundary markers: None present. 3. Capability inventory: Instructional guidance only with no tool calls or system operations. 4. Sanitization: No sanitization of the context file content is performed.
Audit Metadata