dev-kit-review

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection (LOW): The skill ingests untrusted data from ticket files to determine its review steps and verification criteria.
  • Ingestion points: Markdown files located in the .dev-kit/tickets/completed/ directory.
  • Boundary markers: Absent. The instructions do not define clear boundaries between the ticket content and the system instructions, nor do they warn the agent to ignore embedded commands within the tickets.
  • Capability inventory: The skill utilizes powerful tools including run_command (shell execution), view_file (file system access), and agent-browser (web access).
  • Sanitization: No sanitization or validation of the ticket content is performed before the agent uses it to drive these capabilities.
  • Command Execution (SAFE): The skill instructions specify the use of run_command specifically for running tests and linters. While this is a sensitive capability, it is consistent with the primary purpose of a developer review tool and is gated by the instruction 'If safe'.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:37 PM