Frontier AI video generation for agents (Seedance, Runway, Veo, Kling) on your own keys, behind a hard cost cap. Provides MCP tools for text-to-video, image-to-video, and video editing capabilities.
VibeFrame MCP server has 6 tools with basic schemas and descriptions, but significant gaps in parameter documentation and error handling. Tool names follow verb_noun convention (generate_video, edit_video, transcribe_audio, etc.), which is good. However, descriptions are minimal (10-50 chars), parameter descriptions are sparse or missing, and no output schemas are documented. No error recovery guidance. The server exposes external API provider selection (seedance, runway, veo, kling, grok) as free-form string parameters without enum constraints, inviting hallucinated values. File paths are accepted as strings without validation or sanitization guidance. No evidence of idempotency, pagination, or composition patterns.
Analyze video content using Claude or Gemini for scene detection, objects, emotions
Edit video with effects, transitions, and color grading
Generate image from text prompt using AI providers (Gemini, DALL-E, Grok)
Generate speech audio from text using TTS providers (ElevenLabs, OpenAI, Kokoro)
Generate video from text prompt using AI providers (Seedance, Runway, Veo, Kling)
Transcribe audio from video or audio file using Whisper
Provider parameters lack enum constraints. 'provider' accepts free-form strings (seedance, runway, veo, kling, grok) without formal enum declaration, inviting LLM hallucination of invalid provider names.
Tool descriptions are too brief (10-50 chars). Descriptions lack context on WHEN to use each tool, prerequisites, and what the tool returns. LLMs cannot determine tool selection intent from minimal descriptions.
No output schemas documented. Tool responses are not described, forcing LLMs to guess what fields are returned and what downstream tools can chain from these results.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 44 | 2026-07-28+ | v2 |
File path parameters (videoPath, mediaPath) lack validation guidance. No description of expected format, allowed directories, or sanitization against path traversal attacks.
No error handling or recovery guidance. Tools do not document what errors are retryable, what user actions are needed, or how to recover from failures (e.g., API rate limits, invalid provider credentials).