shunk031-manage-public-private-dotfiles
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill is designed to access and manage sensitive configuration files, specifically targeting
~/.local/share/chezmoi-privateand~/.agents/AGENTS-private.md. These paths are identified as containing credentials, authentication helpers, and internal infrastructure settings. While this represents a data exposure surface, the skill includes explicit mitigation instructions to keep credentials and secret values out of command output, diffs, and chat summaries. - [INDIRECT_PROMPT_INJECTION]: The workflow involves reading and following instructions from
AGENTS.mdandAGENTS-private.mdfiles within dotfile repositories. Because these files influence agent behavior and the skill has the capability to write to the filesystem and execute version control commands, there is a surface for indirect prompt injection if those files contain malicious instructions. The skill addresses this by establishing clear ownership boundaries and source-of-truth validation steps.
Audit Metadata