Read-only MCP tools for Seedance-oriented video prompts, reference sequences, diagnosis, and See Dance 2 resource discovery.
Four tools with complete schemas and descriptions, but descriptions are verbose (200+ chars) and lack LLM-optimized brevity. All tools are read-only (good for safety), but naming could be more action-oriented. Schemas are well-structured with proper types, enums, and constraints. No error handling guidance or recovery paths documented. Output schemas are partially visible but lack explicit field documentation.
Turn a rough idea into a deterministic English or Chinese prompt with separate subject and camera motion, exact reference bindings, and continuity controls.
Check English or Chinese prompts for visible motion, camera direction, lighting, continuity, and preserved reference tokens.
Return canonical See Dance 2 prompting, workflow, reference, or safety resources without fetching external content.
Create a deterministic 1–6 shot sequence with exact reference bindings, timing, continuity anchors, and end frames.
Tool descriptions exceed 200 chars and lack LLM-optimized brevity. Baseline is 10 - 1024 chars with optimal ~50 - 200 chars for clarity. Current descriptions bury key intent in verbose prose.
No output schema documentation visible. Tools return structured objects (prompt, shots, resources) but field types and required fields are not declared in the tool definition.
No error handling or recovery guidance. Tools do not document what happens on invalid input (e.g., idea > 5000 chars, shotCount > 6) or how LLMs should respond to failures.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 56 | 2026-07-28+ | v2 |
Naming lacks action verbs. 'get_see_dance_2_resources' is acceptable, but 'build_seedance_prompt' and 'plan_reference_sequence' are imperative rather than verb_noun. Consider 'create_seedance_prompt' or 'generate_seedance_prompt' for consistency.
Parameter descriptions lack constraint details. E.g., 'idea' has minLength/maxLength in schema but description does not state '1 - 5000 characters'. LLMs cannot read JSON Schema, they rely on description text.