email-ops
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and extract information from existing email threads (Step 2) and interpolate that information into LLM prompts for drafting replies (Step 4). This creates a vulnerability where malicious instructions embedded in incoming emails could influence the agent's behavior during the email drafting process.
- Ingestion points:
SKILL.mdStep 2 (Reading existing threads/latest email content via shell or browser). - Boundary markers: Absent. The skill does not define clear delimiters or instructions to ignore embedded commands in the processed email text.
- Capability inventory:
email-opswrapper (SMTP send functionality), File read access (for attachments and body-file). - Sanitization: Absent. There are no explicit validation or filtering steps defined for the external content before it is processed by the LLM.
Audit Metadata