emails
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of markdown instructions and templates. No executable scripts (Python, JavaScript, or shell commands) are included, precluding direct code execution or persistence risks.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing untrusted or external data. 1. Ingestion points: Reads project context from
.agents/product-marketing.mdand processes user-supplied inputs for email content inSKILL.md. 2. Boundary markers: None explicitly defined for isolating content from marketing context files. 3. Capability inventory: The skill is limited to text generation and lacks capabilities for network operations, file writing, or command execution across all reference files andSKILL.md. 4. Sanitization: No specific sanitization or filtering is performed on ingested data. - [DATA_EXPOSURE]: The skill instructions advise the agent to read project-specific marketing context files (e.g.,
product-marketing.md). These files are intended for context-sharing within the environment and do not target sensitive system credentials, private keys, or environment variables.
Audit Metadata