email-write

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it processes instructions and data from external sources without sufficient safeguards.
  • Ingestion points: The workflow dynamically reads content from email-profile.md (user preferences) and references/copy-frameworks.md (framework logic and examples) to determine agent behavior and output structure.
  • Boundary markers: The instructions do not define boundary markers (e.g., XML tags or delimiters) or provide "ignore embedded instructions" warnings when interpolating data from these files into the agent's context.
  • Capability inventory: The skill is granted Read, Write, Grep, and Glob tools. If the ingested files contain malicious instructions, an attacker could potentially trick the agent into misusing these file system capabilities.
  • Sanitization: There is no evidence of sanitization, validation, or filtering of the content retrieved from the local markdown files before it is used to influence the agent's writing process.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 03:04 PM