dispatching-parallel-agents
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill facilitates an attack surface for indirect prompt injection (Category 8) by advising the agent to create new tasks for sub-agents using potentially untrusted data from error logs.
- Ingestion points: Error messages and test failure logs from the environment, as referenced in the 'Agent Prompt Structure' section.
- Boundary markers: The skill lacks recommendations for using delimiters (such as XML tags or triple backticks) or specific instructions for sub-agents to disregard instructions embedded within these logs.
- Capability inventory: Sub-agents are expected to edit production and test code ('Fix bugs in abort implementation') and execute tests, providing a significant capability for an injection to exploit.
- Sanitization: The skill does not provide instructions for sanitizing, escaping, or validating the content of the test logs before they are passed into the sub-agent prompt context.
Audit Metadata