baoyu-imagine

Pass

Audited by Gen Agent Trust Hub on May 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill interfaces with well-known and trusted AI service providers (including OpenAI, Google, Azure, and Alibaba) using authenticated API requests. No malicious network activity or data exfiltration to untrusted domains was detected.
  • [COMMAND_EXECUTION]: Script execution is limited to the skill's own local TypeScript files via bun or npx. The Google provider module utilizes curl for proxy support via a safe argument array in scripts/providers/google.ts, which prevents shell injection vulnerabilities.
  • [PROMPT_INJECTION]: The skill processes untrusted input through prompt files and reference images, creating an indirect prompt injection surface. This is a low-risk factor inherent to the skill's purpose as an AI generator. 1. Ingestion points: Prompts are read from CLI, stdin, and local files (--promptfiles); images are read via --ref (scripts/main.ts). 2. Boundary markers: Absent. 3. Capability inventory: Subprocess calls (curl in google.ts), file-write (scripts/main.ts), network ops (all providers). 4. Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
May 21, 2026, 07:45 AM
Security Audit — agent-trust-hub — baoyu-imagine