advise

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXPOSURE] (SAFE): The skill accesses project-specific documentation (tasks, plans, specs, ADRs) using read-only tools (Read, Glob, Grep). This is necessary for its function as a guidance tool and does not involve exfiltration or exposure of sensitive system credentials.
  • [COMMAND_EXECUTION] (SAFE): No command execution tools are provided to the skill. It is designed for the user to write and execute code manually, while the AI provides advice.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The skill ingests untrusted data from project files (e.g., TASK.md).
  • Ingestion points: ideas/[project]/issues/###-*/TASK.md, PLAN.md, and SPEC-###.md.
  • Boundary markers: None explicitly defined in the provided markdown.
  • Capability inventory: Limited to Read, Glob, Grep, and WebSearch.
  • Sanitization: None detected. The risk is minimized because the agent cannot write files or execute commands based on potentially malicious content in these files.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:29 PM