merge-and-branch

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes dynamic context injection syntax (!) in SKILL.md to execute shell commands (gh pr view, git rev-parse, git status) when the skill is loaded. These commands are used to automatically populate the agent's environment with relevant pull request and branch information.
  • [PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection by ingesting untrusted data from external sources and interpolating it into the agent context.
  • Ingestion points: PR metadata is retrieved via the gh pr view command and injected into the <current_pr> block in SKILL.md.
  • Boundary markers: Absent; the skill does not use specific delimiters or instructions to treat the injected content as untrusted data.
  • Capability inventory: The skill has the capability to execute gh pr merge, git checkout, and git pull based on the processed state.
  • Sanitization: Absent; the skill does not perform validation or sanitization on the content retrieved from the GitHub CLI.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 08:20 PM
Security Audit — agent-trust-hub — merge-and-branch