novita-docs
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches supplemental reference documentation from the author's GitHub repository (novitalabs/novita-skills) to provide up-to-date guidance.
- [COMMAND_EXECUTION]: Instructions guide users to install and use vendor-specific libraries (novita-sandbox) and CLI tools (novita-sandbox-cli) for interacting with the platform.
- [PROMPT_INJECTION]: The skill utilizes a remote-loading architecture for its documentation, which introduces an indirect prompt injection surface.
- Ingestion points: SKILL.md directs the agent to fetch markdown files from raw.githubusercontent.com/novitalabs/novita-skills/.
- Boundary markers: No specific delimiters or safety instructions are used when interpolating the fetched content into the prompt context.
- Capability inventory: The skill facilitates the generation of API requests, SSH commands, and code execution blocks for the user.
- Sanitization: The skill relies on the content hosted in the remote repository and does not perform secondary validation of the fetched text.
Audit Metadata