setup-oz-automations
Pass
Audited by Gen Agent Trust Hub on Jul 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches the
oz-agent-skillfrom thewarpdotdevrepository to enable CLI functionality. - [EXTERNAL_DOWNLOADS]: Installs the
obsidian-headlesspackage from the NPM registry as part of the environment provisioning process. - [CREDENTIALS_UNSAFE]: Handles an Obsidian Sync authentication token; the skill instructs the user to store this sensitive data using the
oz secret createcommand, ensuring it is managed securely by the platform rather than being exposed in logs or prompt history. - [COMMAND_EXECUTION]: Runs a series of setup commands in a Docker environment to configure the vault sync, including creating directories and writing the auth token to a local configuration file.
- [PROMPT_INJECTION]: The skill establishes a pipeline for processing content from an Obsidian vault. This ingestion point represents an indirect prompt injection surface if the vault files contain malicious instructions meant to influence the agent's behavior during scheduled runs.
Audit Metadata