free-tools
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of instructional markdown and reference materials, with no scripts, binaries, or automation components provided.
- [INDIRECT_PROMPT_INJECTION]: The skill creates a data ingestion surface by directing the agent to read project-specific configuration files for business context.
- [INDIRECT_PROMPT_INJECTION]: Ingestion points: Instructions in SKILL.md direct the agent to read .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md.
- [INDIRECT_PROMPT_INJECTION]: Boundary markers: No delimiters or explicit instructions to disregard commands within context files are present.
- [INDIRECT_PROMPT_INJECTION]: Capability inventory: The skill does not define any custom tools, network operations, file-writing capabilities, or shell execution paths.
- [INDIRECT_PROMPT_INJECTION]: Sanitization: No content validation or sanitization logic is implemented for the data read from the local context files.
Audit Metadata