skills/am-will/swarms/parallel-task/Gen Agent Trust Hub

parallel-task

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to the lack of sanitization when processing external markdown plan files.
  • Ingestion points: The skill parses user-provided plan files (e.g., plan.md) in Step 2 to extract task names, descriptions, and acceptance criteria.
  • Boundary markers: The Task Prompt Template in SKILL.md does not use delimiters or explicit instructions to ignore commands or overrides embedded within the task descriptions.
  • Capability inventory: Subagents launched by this skill are granted the ability to read, edit, and commit files to a Git repository as specified in the subagent instructions in Step 3.
  • Sanitization: There is no evidence of sanitization, escaping, or validation of the plan file content before it is interpolated into the prompts for the subagents.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 06:52 AM