dispatching-parallel-agents
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is instructional documentation that defines a 'pattern' for managing concurrent tasks. It does not contain executable code, obfuscation, or sensitive data access.
- [PROMPT_INJECTION]: The skill describes a workflow that captures external data (test names and failure logs) and interpolates them directly into prompts for sub-agents. This presents a surface for indirect prompt injection if the ingested error logs contain malicious instructions.
- Ingestion points: Local test failure logs and error descriptions from standard output.
- Boundary markers: Absent. The recommended prompt templates do not include delimiters or instructions to ignore embedded commands in the test output.
- Capability inventory: The skill uses the
Task()tool to spawn new agents and shell commands (pnpm turbo run test) for verification. - Sanitization: No sanitization or validation of the external test data is mentioned before it is passed to sub-tasks.
Audit Metadata