prioritize

Pass

Audited by Gen Agent Trust Hub on Nov 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from the project context to influence its logic and tool execution.
  • Ingestion points: The skill reads planned work items (tasks, features, stories) from the tasks/ directory or user-provided inline lists as defined in Phase 0.
  • Boundary markers: There are no defined boundary markers or explicit instructions for the AI to ignore potential instructions embedded within the task data being prioritized.
  • Capability inventory: The skill has the capability to call the $task-track tool and modify the frontmatter of task files to record priority and priority_label. It can also initiate an $llm-council involving 11 sub-agents.
  • Sanitization: The workflow does not specify any sanitization or validation of the content within the work items, meaning malicious instructions in a task description could potentially influence the scoring results or downstream tool calls.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 4, 2026, 10:12 AM