parallel-task-spark
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to parse and execute logic defined in external markdown plan files, which serves as a vector for indirect instructions.
- Ingestion points: The process described in Step 2 ("Read & Parse Plan") involves reading task descriptions, dependencies, and validation criteria directly from user-provided plan files (e.g.,
plan.md). - Boundary markers: The instructions lack specific boundary markers or strict delimiters to differentiate between the agent's instructions and potentially malicious commands embedded within the plan data.
- Capability inventory: The skill orchestrates subagents (
sparkyrole) with the capability to perform file system modifications, write and execute code (TDD tests), and run validation commands based on the extracted plan content. - Sanitization: There is no evidence of sanitization or validation logic to ensure that instructions extracted from the markdown files do not override the agent's intended behavior or security constraints.
Audit Metadata