doctor
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a series of shell commands, including
find,ls,rg, andbasename, along with the specializedomxCLI to perform diagnostics of the installation environment and configuration files. - [EXTERNAL_DOWNLOADS]: Retrieves version information for the
oh-my-codexpackage from the official npm registry. This is used to inform the user of available updates during the diagnostic process. - [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from local configuration files and command outputs, creating a vulnerability surface for indirect prompt injection.
- Ingestion points: The skill reads local configuration files such as
config.toml,settings.json, andAGENTS.md, as well as scripts located in the hooks directory. - Boundary markers: Detailed instructions define a specific "Report contract" and remediation logic to ensure the agent processes findings in a controlled manner.
- Capability inventory: The agent is capable of modifying environment settings and tool hooks through the
omx setupcommand. - Sanitization: The instructions mandate the use of preview modes (
--dry-run) and forbid the deletion of non-OMX files, requiring the agent to record exact targets before any modification. - [SAFE]: All diagnostic and remediation activities are scoped to the
oh-my-codextool. The skill provides explicit instructions to prevent accidental data loss and ensures that all repairs are performed within a strictly authorized scope.
Audit Metadata