product-marketing-context
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill scans local repository files (such as README, landing pages, and package.json) to auto-draft marketing documentation. This ingestion of untrusted data creates a potential surface for indirect prompt injection if those files contain malicious instructions.
- Ingestion points: The skill scans various files in the local codebase as defined in SKILL.md.
- Boundary markers: The instructions do not define specific delimiters or warnings for the agent to ignore instructions embedded in the ingested repository content.
- Capability inventory: The agent has the capability to write the generated context to the filesystem at .agents/product-marketing-context.md.
- Sanitization: No explicit sanitization or filtering of codebase content is specified before the document is generated.
- Mitigation: The workflow includes a review step where the agent must show the completed document to the user and ask for adjustments before saving, which acts as a human-in-the-loop security control.
Audit Metadata