parallel-agents
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE] (SAFE): No malicious patterns detected in the skill instructions or code snippets.
- [COMMAND_EXECUTION] (LOW): The skill suggests the use of
mkdir -pto create local planning directories. This is a standard, low-risk filesystem operation intended for project organization. - [PROMPT_INJECTION] (LOW): The skill is susceptible to indirect prompt injection (Category 8) during the aggregation phase.
- Ingestion points: The orchestrator reads
findings.mdandprogress.mdgenerated by sub-agents in Step 5. - Boundary markers: Present; the skill recommends using agent/task headings to delimit content.
- Capability inventory: The skill uses a
Task()tool to dispatch sub-agents and performs local file writes/appends. - Sanitization: Absent; the instructions do not explicitly mention sanitizing or validating the content provided by sub-agents before aggregation.
Audit Metadata