pull-request

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • PROMPT_INJECTION (LOW): The skill uses 'IMPORTANT' and negative constraints like 'Do NOT check git status' to override the agent's default safety and helpfulness behaviors.
  • COMMAND_EXECUTION (LOW): The skill uses shell interpolation $(git branch --show-current) within a gh command. While common in CLI tools, this is a potential injection surface if local branch names are controlled by an untrusted party.
  • INDIRECT_PROMPT_INJECTION (LOW): The skill ingests data from existing pull requests (titles and descriptions) and GitHub CLI output without explicit boundary markers or sanitization.
  • Ingestion points: GitHub CLI output and existing PR content.
  • Boundary markers: Absent.
  • Capability inventory: PR creation and modification via delegated GitHub CLI operations.
  • Sanitization: None detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:22 PM