multi-agent-task-orchestrator
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted task descriptions which serves as an attack surface for indirect prompt injection. 1. Ingestion points: The check_duplicate and route_task functions in SKILL.md ingest the description parameter. 2. Boundary markers: No delimiters or ignore instructions are specified for the input data to prevent the agent from following instructions embedded in descriptions. 3. Capability inventory: The orchestrator is instructed to use shell commands like git, npm, and pytest. 4. Sanitization: No input filtering or sanitization is implemented for the processed descriptions.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to verify the work of specialized agents. Evidence includes the use of git diff, npm test, pytest, grep, and npm run build within the quality gate instructions.
Audit Metadata