email-drafting

Pass

Audited by Gen Agent Trust Hub on Mar 19, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [NO_CODE]: The skill is composed strictly of Markdown documentation, guidelines, and examples. It does not contain scripts, configuration files, or binaries.
  • [PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection through the ingestion of untrusted user data.
  • Ingestion points: The Usage section in SKILL.md identifies recipient details, purpose, and key points as user-controlled input fields.
  • Boundary markers: There are no instructions for the agent to use delimiters (such as XML tags or triple quotes) to encapsulate user input, nor are there instructions to ignore instructions embedded within that input.
  • Capability inventory: The skill enables the agent to generate complete email content (Subject, Body, and Signature) for external communication.
  • Sanitization: The instructions do not include any steps for the agent to sanitize or validate the content of the provided input before incorporating it into the final draft.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 19, 2026, 08:23 AM