git-conflict-resolve
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core function of reading and analyzing code from conflicted files.
- Ingestion points: The agent reads three-way content of conflicted files using
git show :1:/:2:/:3:<FILE>in Sub-stage Y.2 and compares resolution results usinggit showin Sub-stage Y.5. - Boundary markers: The instructions do not define clear delimiters (e.g., XML tags or unique markers) for the content read from files, nor do they include explicit warnings for the agent to ignore any instructions found within that content.
- Capability inventory: The skill has significant capabilities including executing shell commands for repository manipulation (
git merge,git rebase,git commit,git add) and file system operations (git rm -rfq <DIR>). - Sanitization: There is no evidence of sanitization or escaping of the file content before it is processed by the AI agent's semantic analysis engine.
Audit Metadata