product-marketing-context
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The auto-draft feature reads and processes content from repository files like README.md and landing pages. This creates a surface for indirect prompt injection where malicious instructions in those files could influence the agent's behavior while drafting.
- Ingestion points: Reads various local repository files (SKILL.md, Step 2).
- Boundary markers: The skill lacks specific instructions for the agent to treat the analyzed content strictly as data rather than instructions.
- Capability inventory: The agent has permissions to read the repository and write the resulting context file to the .agents/ directory.
- Sanitization: No sanitization or filtering is performed on the content gathered from the codebase.
Audit Metadata