workspace-settings
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill acts as a documentation guide for an AI agent to interact with the Windmill platform's workspace settings. It does not contain any executable scripts, obfuscated code, or malicious instructions.
- [COMMAND_EXECUTION]: The skill documents the use of the wmill CLI for administrative tasks. These operations are restricted to workspace admins and are part of the legitimate functionality of the Windmill platform.
- [INDIRECT_PROMPT_INJECTION]: The skill manages workspace configurations that could serve as ingestion points for external data, representing a surface for indirect prompt injection.
- Ingestion points: Workspace configuration retrieved via wmill workspace get-settings or settings.yaml.
- Boundary markers: No specific delimiters are provided in the documentation to isolate potentially untrusted input.
- Capability inventory: Administrative reading and writing of workspace settings via the wmill CLI.
- Sanitization: The documentation explicitly states that all write operations are subject to server-side validation and audit logging.
- [SAFE]: The skill includes security-conscious instructions for handling sensitive data, specifically utilizing the SENSITIVE_AND_UNCHANGED placeholder to prevent accidental exposure or overwriting of credentials.
Audit Metadata