quality-gate

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • PROMPT_INJECTION (LOW): This skill is vulnerable to Indirect Prompt Injection (Category 8) as it processes untrusted data from external sources.
  • Ingestion points: Processes content from issue/PR links, Requirements in scope, and Files changed / diff.
  • Boundary markers: The instructions do not define clear delimiters or escaping mechanisms to prevent the LLM from following instructions embedded within the code or requirements documents being audited.
  • Capability inventory: The skill orchestrates multiple specialist passes including ubs-scanner, security-audit, and execution of project-specific test/build commands.
  • Sanitization: There is no evidence of sanitization or validation of the external content before it is processed by the agent or passed to downstream tools.
  • COMMAND_EXECUTION (LOW): The workflow explicitly instructs the agent to "Prefer running the project’s actual commands" to record evidence. While this is the intended primary purpose of a verification lead skill, it creates a risk where a malicious PR could modify local build scripts or test configurations to execute arbitrary code when the agent attempts to run the 'actual commands' for validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:36 PM