parallel-task-spark
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon data from external markdown plan files, which is an inherent part of its orchestration function.
- Ingestion points: Processes markdown files (e.g.,
plan.md) to extract task details, dependencies, and validation instructions. - Boundary markers: The instructions do not explicitly define delimiters or include specific warnings to ignore embedded instructions within the ingested plan files.
- Capability inventory: The skill manages subagents capable of codebase research, file modification, shell command execution for testing/validation, and Git commits.
- Sanitization: No explicit sanitization of task content or validation commands is performed prior to processing or subagent delegation.
Audit Metadata