promptflow-webmcp.netlify.app WebMCP Tools, Schemas & Agent Capabilities

← WebMCP Directory

promptflow-webmcp.netlify.app

live Unverified (seed import) AI & Agents 49 tools imperative implementation

Open website How verification works

PromptFlow.ai

First seen 2026-11-05 · Last seen 2026-11-05 · Source: webmcp.com

Answer 7 · Act 42 · Transact 0 · Score: Not yet scored

Tool inspector

build_and_verify_system

act imperative

RECOMMENDED ONE-CALL WebMCP workflow. Turn an arbitrary human software brief into a live architecture and a verified generated project. The calling AI agent should infer semantic components/connections when needed; PromptFlow validates and applies the plan, performs deterministic structural repair and scaling, then runs the full Judge Mode pipeline: architecture intelligence, digital-twin failure injection, hardening, re-test, code generation, real WebContainer install/test/build/runtime verification, production QA, and final verdict. Use this instead of chaining many tools when the user asks to create and prove a complete system.

Input schema

Property	Type	Required	Description
`request`	string	yes	The user's complete product/system brief. Preserve important domain, scale, realtime, security, data, AI, media, and integration constraints.
`targetUsers`	number	no	Optional explicit user-scale target. Omit when the brief contains no meaningful scale requirement.
`replaceCurrent`	boolean	no	Replace the current canvas before building. Defaults to true. Set false only when ids are guaranteed not to conflict with existing nodes.
`autoFix`	boolean	no	Automatically repair deterministic structural findings after the initial build. Defaults to true.
`components`	array	no	Optional agent-authored architecture plan. For arbitrary briefs, provide the components you infer semantically so PromptFlow can apply the whole plan in one tool call. Omit to use PromptFlow's local intent compiler fallback.
`connections`	array	no	Relationships between the optional agent-authored components.

Raw JSON schema

{
  "type": "object",
  "properties": {
    "request": {
      "type": "string",
      "maxLength": 2400,
      "description": "The user's complete product/system brief. Preserve important domain, scale, realtime, security, data, AI, media, and integration constraints."
    },
    "targetUsers": {
      "type": "number",
      "description": "Optional explicit user-scale target. Omit when the brief contains no meaningful scale requirement."
    },
    "replaceCurrent": {
      "type": "boolean",
      "description": "Replace the current canvas before building. Defaults to true. Set false only when ids are guaranteed not to conflict with existing nodes."
    },
    "autoFix": {
      "type": "boolean",
      "description": "Automatically repair deterministic structural findings after the initial build. Defaults to true."
    },
    "components": {
      "type": "array",
      "maxItems": 40,
      "items": {
        "type": "object",
        "properties": {
          "id": {
            "type": "string",
            "maxLength": 120,
            "description": "Stable unique component id using lowercase kebab-case when possible."
          },
          "label": {
            "type": "string",
            "maxLength": 160,
            "description": "Human-readable software component name."
          },
          "type": {
            "type": "string",
            "enum": [
              "frontend",
              "backend",
              "api",
              "database",
              "cache",
              "queue",
              "payment",
              "auth",
              "service",
              "external",
              "gateway",
              "worker",
              "cdn",
              "observability"
            ]
          },
          "description": {
            "type": "string",
            "maxLength": 800,
            "description": "Concise responsibility and architectural purpose."
          }
        },
        "required": [
          "id",
          "label",
          "type",
          "description"
        ],
        "additionalProperties": false
      },
      "description": "Optional agent-authored architecture plan. For arbitrary briefs, provide the components you infer semantically so PromptFlow can apply the whole plan in one tool call. Omit to use PromptFlow's local intent compiler fallback."
    },
    "connections": {
      "type": "array",
      "maxItems": 80,
      "items": {
        "type": "object",
        "properties": {
          "sourceId": {
            "type": "string",
            "maxLength": 120
          },
          "targetId": {
            "type": "string",
            "maxLength": 120
          },
          "label": {
            "type": "string",
            "maxLength": 160
          }
        },
        "required": [
          "sourceId",
          "targetId"
        ],
        "additionalProperties": false
      },
      "description": "Relationships between the optional agent-authored components."
    }
  },
  "required": [
    "request"
  ],
  "additionalProperties": false
}