repair-repository-integrity
Pass
Audited by Gen Agent Trust Hub on Jul 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: Analysis of the skill instructions and reference documents found no evidence of malicious patterns, obfuscation, or unauthorized data access.
- [COMMAND_EXECUTION]: The skill utilizes Git commands for repository repair. It proactively includes security instructions to mitigate command injection risks by mandating the use of
--end-of-optionsand preventing shell interpolation of repository-sourced strings. - [PROMPT_INJECTION]: The skill identifies Git repository metadata such as commit messages and configurations as untrusted content in
SKILL.md. It addresses potential indirect prompt injection by instructing the agent to treat this content as data rather than authority and to specifically ignore any embedded instructions. This is further reinforced by recommending the use of boundary markers like--end-of-optionsfor all command invocations.
Audit Metadata