UNPKG

comfyui-mcp

Version:

Local-first, agent-native control plane for ComfyUI — MCP server + autonomous sidebar agent that drives your live graph in natural language on ANY LLM: Claude/ChatGPT/Gemini on your subscription (no API key), free local models via Ollama (fully offline),

61 lines • 3.06 kB
import { z } from "zod"; import { generateImage } from "../services/generate-image.js"; import { enqueueWorkflow } from "../services/workflow-executor.js"; import { listLocalModels } from "../services/model-resolver.js"; import { errorToToolResult } from "../utils/errors.js"; async function resolveCheckpoint() { try { const models = await listLocalModels("checkpoints"); return models[0]?.name; } catch { return undefined; } } export function registerGenerateImageTool(server) { server.tool("generate_image", "Generate an image from a text prompt — the high-level entry point. Builds a txt2img workflow, " + "filling any unspecified parameter from your configured defaults (set_defaults / COMFYUI_DEFAULT_* / config file), " + "auto-selecting a local checkpoint when none is given. Returns the prompt_id immediately; the resulting " + "asset_id arrives in the completion notification and can be passed to view_image or regenerate. " + "For full control over the node graph, use create_workflow + enqueue_workflow instead.", { prompt: z.string().describe("Positive text prompt"), negative_prompt: z.string().optional().describe("Negative prompt (default: empty / from defaults)"), width: z.number().int().positive().optional().describe("Image width"), height: z.number().int().positive().optional().describe("Image height"), steps: z.number().int().positive().optional().describe("Sampling steps"), cfg: z.number().positive().optional().describe("CFG scale"), sampler: z.string().optional().describe("Sampler name (e.g. euler, dpmpp_2m)"), scheduler: z.string().optional().describe("Scheduler (e.g. normal, karras)"), seed: z.number().int().optional().describe("Seed (omit to randomize)"), checkpoint: z .string() .optional() .describe("Checkpoint filename; auto-selected from local models if omitted"), batch_size: z.number().int().positive().optional().describe("Number of images to generate"), }, async (args) => { try { const result = await generateImage(args, { resolveCheckpoint, enqueue: (workflow) => enqueueWorkflow(workflow), }); return { content: [ { type: "text", text: JSON.stringify({ status: "enqueued", prompt_id: result.prompt_id, queue_remaining: result.queue_remaining, checkpoint: result.checkpoint, note: "asset_id will be available in the completion notification; use view_image or regenerate with it.", }, null, 2), }, ], }; } catch (err) { return errorToToolResult(err); } }); } //# sourceMappingURL=generate-image.js.map