run-parallel-agents-feature-debug
Pass
Audited by Gen Agent Trust Hub on Mar 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the processing of external debugging data. * Ingestion points: SKILL.md (Step 3: Prepare Debugging Briefs) instructs the agent to gather error messages and stack traces from the environment or user and include them in the prompt parameter for sub-agents. * Boundary markers: Absent; the skill lacks mandatory boundary markers or instructions for the sub-agent to ignore potential commands within the error data. * Capability inventory: SKILL.md uses the Agent tool to launch sub-agents which possess filesystem modification and command execution capabilities. * Sanitization: Absent; there is no mention of sanitization, escaping, or filtering of external content before it is interpolated into sub-agent prompts.
Audit Metadata