orchestrate
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates as an orchestrator that ingests results and reports from subagents, which represents a potential surface for indirect prompt injection. \n
- Ingestion points: Subagent reports, file diffs, and test outputs are ingested into the agent context (SKILL.md). \n
- Boundary markers: The skill does not define specific text-based delimiters for isolating subagent content. \n
- Capability inventory: The agent is authorized to run verification commands and relevant tests or builds (SKILL.md). \n
- Sanitization: The risk is mitigated by explicit instructions to 'treat every self-report as unverified' and to perform independent verification via tests and file inspections. \n- [SAFE]: No other security concerns were identified. The skill does not contain hardcoded credentials, malicious network operations, obfuscation, or privilege escalation attempts. The instructions focus on establishing a controlled and verifiable workflow.
Audit Metadata