messages-ops
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process text from external messaging platforms (iMessage, Social DMs), which serves as an entry point for indirect prompt injection. A malicious message could contain instructions intended to influence the agent's behavior during inspection or follow-up drafting.
- Ingestion points: retrieval of live messages from local surfaces, social DMs, and other browser-gated surfaces (SKILL.md).
- Boundary markers: The skill does not define delimiters or provide specific instructions to ignore potentially malicious content within the retrieved messages.
- Capability inventory: The workflow involves reading inbound messages, identifying open loops, and handing off to intelligence or outbound skills (SKILL.md).
- Sanitization: No content filtering, validation, or sanitization requirements are specified for the message data retrieved from external sources.
Audit Metadata