dino-manage-tags
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to executedinoCLI commands for listing, creating, and modifying tags. This is the intended functionality. - [DATA_EXPOSURE]: The skill follows security best practices for secret management by instructing the agent not to ask for tokens and to rely on environment variables (
DINOX_TOKEN) or the CLI's own authentication mechanism. - [PROMPT_INJECTION]: The skill identifies an indirect prompt injection surface when processing user-provided tag names.
- Ingestion points: User-supplied tag names provided via the
argument-hintand input parameters. - Boundary markers: The instructions explicitly warn to treat user-provided tag names as untrusted and to avoid running non-
dinoshell commands. - Capability inventory: The skill uses the
Bashtool for command execution. - Sanitization: The skill requires showing the exact command to the user and getting explicit confirmation before any write operations, which acts as a manual review checkpoint.
Audit Metadata