paw-mkt-product-context
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local setup scripts (tool-discovery.sh) to verify the availability of required development tools like Node.js and Python within the project environment.
- [EXTERNAL_DOWNLOADS]: It provides instructions for downloading and installing the agent-browser tool from a trusted GitHub organization (Vercel Labs), which is used for market research.
- [COMMAND_EXECUTION]: The skill instructs the agent to use browser automation via the agent-browser CLI for research, providing clear guidance on managing authentication and session security.
- [PROMPT_INJECTION]: The skill ingests data from local strategy documents (SOSTAC files) to populate marketing templates. While this creates an indirect prompt injection surface, it is consistent with the skill's primary function of processing project context.
- Ingestion points: Reads local SOSTAC and brand context files in markdown format.
- Boundary markers: Uses structured templates for output but does not define explicit escaping or sanitization for ingested content.
- Capability inventory: File system write operations and shell command execution via browser automation tools.
- Sanitization: No content filtering is performed on ingested data before processing.
Audit Metadata