scramble-rotator
Pass
Audited by Gen Agent Trust Hub on Jul 6, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE]: The skill references a sensitive file path at
~/.build-host/credentials.json. This is used to authenticate with the build.host API. The skill includes specific safety instructions to prevent the exposure or exfiltration of these credentials, such as forbidding the printing of API keys or tokens in chat logs or source code. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the user's project by reading the 'current UI and constraints'. This is an attack surface where malicious content in a project's HTML or CSS could influence agent behavior, though the skill's narrow focus on design tokens and CSS effects minimizes the risk.
- [COMMAND_EXECUTION]: The skill performs file system edits and network requests to the build.host API (
POST /api/projects/:uuid/deploy) to facilitate project updates and deployments. These actions are within the intended scope of the skill for the specified platform.
Audit Metadata