repo-health
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses local shell commands (bash) to audit the repository state, including checking for file existence, git configuration, and tool versions (e.g.,
git config core.hooksPath,bd --version,timbers doctor). - [DATA_EXPOSURE]: The skill includes a dedicated check for secrets in tracked files (
.env,.pem,.key, etc.). It classifies this as a critical finding for the user and explicitly instructs the agent to warn the user rather than attempting an automated fix, preventing accidental data mishandling. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data by reading repository configuration files (e.g.,
package.json,tsconfig.json,.golangci.yml). This is a necessary part of its auditing function, and the skill implements human-in-the-loop safeguards by requiring confirmation before acting on any findings. - [SAFE]: No network exfiltration, obfuscation, or unauthorized remote code execution patterns were detected. All external tools mentioned (such as
bd,timbers, anddolt) are referenced in the context of standard development environment setup.
Audit Metadata