Skills
Navigate
skills/tanstack/ai/ai-core/media-generation/Gen Agent Trust Hub

ai-core/media-generation

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a comprehensive guide for media generation (images, audio, video, speech, transcription) using the TanStack AI SDK.
  • [EXTERNAL_DOWNLOADS]: The skill references official TanStack packages, including @tanstack/ai, @tanstack/ai-react, and provider-specific adapters like @tanstack/ai-openai and @tanstack/ai-gemini. These are legitimate vendor-owned resources.
  • [INDIRECT_PROMPT_INJECTION]: The skill demonstrates handling user-provided prompts for generating media. It includes examples of using server-side input validation (e.g., inputValidator in TanStack Start) to ensure data integrity. The capabilities are scoped to media generation via official provider APIs.
  • Ingestion points: User-supplied prompt, text, turns, and audio (base64/data URLs) are processed in SKILL.md.
  • Boundary markers: Boundary markers are not explicitly mentioned for media generation prompts, which is typical for this domain.
  • Capability inventory: Perform network operations (API calls) to media providers (OpenAI, Gemini, fal.ai, etc.) and file/blob storage via persistence middleware.
  • Sanitization: Recommends inputValidator for server functions to validate and sanitize incoming data.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 04:19 AM