resolving-merge-conflicts
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to discover and run "automated checks" such as typechecking, tests, and formatting tools (SKILL.md, step 4). It also involves executing git commands to manage the merge and rebase process (SKILL.md, steps 1, 5). These are standard operations within a development environment.
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted external data during the conflict resolution process.
- Ingestion points: Conflicting files (SKILL.md, step 1), commit messages, and Pull Request descriptions (SKILL.md, step 2).
- Boundary markers: Standard Git conflict markers (<<<<<<<, =======, >>>>>>>) serve as delimiters in the file content.
- Capability inventory: Execution of shell commands for version control and running project-defined scripts (tests/linters).
- Sanitization: There is no explicit requirement for the agent to sanitize or escape text retrieved from external PR comments or commit history before resolution.
Audit Metadata