resolving-merge-conflicts
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions define a standard process for resolving git conflicts, including analyzing history, resolving code hunks, and running project-specific tests. No suspicious activities, obfuscation, or safety bypasses were found.
- [COMMAND_EXECUTION]: The agent is instructed to run automated checks (typecheck, tests, format) and git commands. These operations are within the scope of the skill's primary purpose for conflict resolution within a local workspace.
- [PROMPT_INJECTION]: The skill processes untrusted data from git history, commit messages, PRs, and issue trackers (Ingestion points). There are no specific instructions or delimiters to ignore embedded instructions (Boundary markers). The agent has the capability to run git commands and project tests (Capability inventory). No explicit sanitization of the processed external data is defined (Sanitization). This represents an indirect prompt injection surface inherent to the task of code maintenance.
Audit Metadata