workflow-preflight

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes specific git and gh commands to gather repository state, check for uncommitted changes, and detect merge conflicts. It uses git merge-tree for dry-run conflict detection, which is a safe practice.
  • [EXTERNAL_DOWNLOADS]: Fetches remote state and prunes stale references from the origin repository on GitHub to ensure the local environment is synchronized.
  • [PROMPT_INJECTION]: The skill ingests metadata from GitHub issues and pull requests, creating a surface for indirect prompt injection.
  • Ingestion points: Issue and PR metadata fetched via gh issue view and gh pr list in SKILL.md.
  • Boundary markers: No markers or explicit instructions are provided to the agent to treat this data as untrusted.
  • Capability inventory: The skill has the ability to execute git/gh commands and perform file operations (Read, TodoWrite).
  • Sanitization: Metadata from the external repository is processed without sanitization or validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 08:12 PM