stack-rebase

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through external git and GitHub data.
  • Ingestion points: Branch names are ingested via git branch --list (Step 3, 5, 6) and PR metadata is retrieved via gh pr list (Step 6).
  • Boundary markers: None. The skill does not include instructions to ignore potentially malicious content within branch names or PR metadata.
  • Capability inventory: The skill utilizes bash, git, and gh (GitHub CLI) across multiple scripts to perform checkouts, rebases, force-pushes, and PR edits.
  • Sanitization: Variable interpolation in shell scripts (e.g., "${branch}") provides basic protection against word splitting but does not fully sanitize against sophisticated command injection if malicious characters are present in the ingested branch names or PR metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 01:39 AM
Security Audit — agent-trust-hub — stack-rebase