workflow-pm-reporting

Pass

Audited by Gen Agent Trust Hub on May 1, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill uses repetitive and forceful language such as "IMPORTANT MUST ATTENTION" and "NEVER batch-complete" to ensure the agent follows a rigid workflow. These instructions are directed at task completion logic and do not aim to override safety protocols or elicit restricted behavior.
  • [PROMPT_INJECTION]: The skill ingests user input to drive a multi-step workflow, creating a potential surface for indirect prompt injection.
  • Ingestion points: The user's prompt is passed as context to the /workflow-start command in SKILL.md.
  • Boundary markers: No delimiters or specific markers are defined to isolate the user input from the skill's instructions.
  • Capability inventory: The skill uses TaskCreate and workflow management tools (/status, /dependency); it does not exhibit dangerous capabilities like arbitrary code execution or file system modification.
  • Sanitization: There is no evidence of sanitization or validation of the input prompt, though the risk is mitigated by the limited scope of the reporting tools.
Audit Metadata
Risk Level
SAFE
Analyzed
May 1, 2026, 05:52 PM