gumlet-video-workspace

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious code, obfuscation, or suspicious activities were identified.
  • [COMMAND_EXECUTION]: The skill uses the 'gumlet' CLI to perform workspace management tasks like listing, creating, and deleting workspaces.
  • [PROMPT_INJECTION]: The skill interpolates user-controlled inputs into shell commands, creating a potential surface for indirect prompt injection. 1. Ingestion points: Workspace IDs and names in SKILL.md. 2. Boundary markers: None. 3. Capability inventory: Command execution via the 'gumlet' CLI. 4. Sanitization: None mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 11:13 AM