baoyu-xhs-images

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill documentation in references/codex-imagegen.md describes a fallback mechanism using npx -y bun to execute local scripts if the Bun runtime is not found on the system. This involves downloading the tool from the npm registry.
  • [DYNAMIC_EXECUTION]: The skill dynamically discovers and executes local scripts or binaries, such as those specified by the BAOYU_CODEX_IMAGEGEN_BIN environment variable or found within the skill's package structure (e.g., packages/baoyu-codex-imagegen/src/main.ts).
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided content to generate image prompts. While it lacks explicit sanitization or boundary markers, the risk is low as the primary output is static images. Evidence: 1. Ingestion points: SKILL.md Step 1 and analysis-framework.md. 2. Boundary markers: Absent. 3. Capability inventory: Skill tool calls, file writing, and script execution. 4. Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 07:47 AM
Security Audit — agent-trust-hub — baoyu-xhs-images