MCP server for generating images on ComfyUI from a chat client
Two tools with clear, actionable descriptions and proper verb-noun naming. list_workflows and generate_image both have detailed docstrings (194+ chars) explaining WHAT, WHEN, and WHY. generate_image has a well-structured input schema with 3 parameters, each with type and description. However, list_workflows has an empty input schema ({}), and neither tool documents output structure. Descriptions are LLM-optimized and include dependency hints ('Call this instead of asking the user'). No parameter validation constraints (enums, ranges) are visible. Error handling guidance is present in descriptions but not formalized in schema.
Generate an image on ComfyUI and return it inline. Returns "<workflow> · seed <seed> · <url>" plus -- unless this server was started with --no-inline-image -- the image itself. Clients usually collapse tool results, so the image is easy to miss. Pass the URL on in your reply: it survives the collapse, and is how the user sees the picture when no image is returned. Write it bare so the client turns it into a link -- code formatting or backticks make it unclickable. Use exactly the workflow name the user asked for. Similar names are different workflows -- "Krea2" and "Krea2+Upscale" are not interchangeable. Fails if the workflow needs an input image or has no prompt node to patch.
List the workflows saved in ComfyUI, ready ones first. Call this instead of asking the user which workflows they have -- they expect you to look. The reply separates workflows that can generate from those that cannot, because the models they reference are not installed or no prompt node was found. Only use the ready ones; naming another will fail.
list_workflows has empty input schema ({}). While the tool takes no parameters, the schema should be explicit and documented.
Output schemas are not documented. Neither tool declares what fields the response contains. LLMs cannot plan downstream calls or extract data without knowing response structure.
generate_image 'seed' parameter lacks validation constraints. Description says 'Reuse a seed from an earlier result' but does not state the valid range (0 to MAX_SEED=2^32-1).
generate_image 'workflow' parameter accepts free-form string. Should accept enum of available workflows from list_workflows output, or at minimum document the exact format expected.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 58 | 2026-07-28+ | v2 |
Error handling is narrative (embedded in descriptions) rather than formalized. No structured error classification (retryable vs user-fixable vs fatal) or recovery guidance in schema.