multi-agent-mode
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where a lead agent ingests and integrates data from multiple helper agents, creating a potential surface for indirect prompt injection if a helper agent processes malicious external content.
- Ingestion points: The lead agent ingests "helper outputs" and "reports" during the integration and verification phase (SKILL.md).
- Boundary markers: There are no explicit structural delimiters or boundary markers specified for the ingested helper data within the instructions (SKILL.md).
- Capability inventory: The lead agent is responsible for "direct execution," "integration," and "final verification," which typically involves file system access and command execution capabilities (SKILL.md).
- Sanitization: The skill mandates that the lead agent must "inspect helper output, diffs, or changed paths before accepting" and run final verification to ensure result integrity (SKILL.md).
Audit Metadata