product-marketing-context
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to read local files (README, package.json, landing pages) to gather information for a marketing document. This is a standard project analysis task and does not involve executing dangerous system commands or accessing sensitive files outside the project scope.- [DATA_EXPOSURE_&_EXFILTRATION]: The skill creates and writes to a file at
.claude/product-marketing-context.md. While it interacts with the file system, these operations are confined to the project directory for the purpose of storing user-provided marketing information. There are no network requests or external data transfers.- [INDIRECT_PROMPT_INJECTION]: The skill processes project files (README, etc.) which could theoretically contain malicious instructions. However, the skill treats this data as informational content for a static document template and does not feed this data into dangerous tools or executable environments. The risk of indirect injection triggering harmful actions is minimal.
Audit Metadata