git-conflict-resolve

Pass

Audited by Gen Agent Trust Hub on Nov 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes conflicted file content during git operations (merge, rebase, cherry-pick). This content originates from external branches or commits and could contain malicious instructions designed to manipulate the agent's resolution logic or report output.
  • Ingestion points: File content is read and analyzed in SKILL.md under 'Step 3b: Read both versions' and 'Step 4: Resolve each conflict'.
  • Boundary markers: The skill relies on standard Git conflict markers (<<<<<<<, =======, >>>>>>>) to delineate versions, which provides structural but not semantic isolation.
  • Capability inventory: The agent can execute git checkout, git add, and continuation commands (git rebase --continue), as well as edit files manually.
  • Sanitization: No explicit sanitization or instruction-filtering is applied to the content of conflicted files before analysis.
  • [COMMAND_EXECUTION]: The skill uses local shell commands to interact with the Git version control system and manage files.
  • Evidence: SKILL.md utilizes git status, git diff, git checkout, git add, and environment variables like GIT_EDITOR=true for non-interactive continuations.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 4, 2026, 10:11 AM