Web Research Agent (Prompt to Structured Data)

omnial/agent

Given a research prompt -- optionally constrained to specific URLs and a maximum spend -- autonomously reads web pages to find and extract the requested data. One call returns a plain-language summary of what it found plus whatever structured fields it identified there, or an empty result when nothing matching the prompt turns up.

activeper callv4

Price

Up to $0.137 per call; you pay what the call used

You pay the price shown. If a call turns out to cost more than its quote, you pay the actual price, capped at twice the quote. No monthly minimum. Top-ups start at $5.00.

Data source

The websites the agent visits

Gathered with a language model from the websites the agent visits for your prompt.

Use it

Your AI assistant does the work: it finds this tool, shows you the price first, runs it, and you are charged when the run finishes. There is nothing to set up for this tool in particular.

Connect once and your AI can find and run this tool, and any other it needs.

Install it Create an account

For developers: the calls your assistant makes

agent

// 1. search: find the tool
omnial_search({ query: "what you need" })

// 2. inspect: price and input schema
omnial_inspect({ tool: "omnial/agent" })

// 3. learn: read the usage guide at runtime
omnial_learn({ tool: "omnial/agent" })

// 4. run: price the exact call first, then run it
omnial_execute({ tool: "omnial/agent", input: { /* see input_schema */ }, dry_run: true })
omnial_execute({ tool: "omnial/agent", input: { /* ... */ } })

// 5. settle: the call is charged when it finishes, nothing to do

Not connected yet? Point your MCP client at https://mcp.eden-engine.com/mcp, then copy a config for your client.

Pricing

Every charge this tool can incur.

Charge	Price
Per call	$0.137

Updated: Oct 27, 2026

Input and output schemas

Input schema

input_schema.json

{
  "type": "object",
  "required": [
    "prompt"
  ],
  "properties": {
    "urls": {
      "type": "array",
      "items": {
        "type": "string",
        "format": "uri"
      },
      "maxItems": 50,
      "description": "Constrain the agent to these URLs."
    },
    "prompt": {
      "type": "string",
      "examples": [
        "Find the pricing tiers and per-seat cost on acme.com"
      ],
      "maxLength": 10000,
      "minLength": 1,
      "description": "What data to find and extract."
    },
    "schema": {
      "type": "object",
      "description": "JSON Schema to shape the extracted data.",
      "propertyNames": {
        "type": "string"
      },
      "additionalProperties": {}
    },
    "maxCredits": {
      "type": "integer",
      "default": 100,
      "maximum": 100,
      "minimum": 1,
      "description": "Maximum provider credits the agent may spend on this task."
    },
    "strictConstrainToURLs": {
      "type": "boolean"
    }
  },
  "additionalProperties": false
}

Output schema

output_schema.json

{
  "type": "object",
  "properties": {
    "data": {
      "type": "object",
      "properties": {
        "plans": {
          "type": "array",
          "items": {
            "type": "object",
            "properties": {
              "name": {
                "type": "string",
                "description": "The plan's own name, e.g. 'Standard' -- not a person's name."
              },
              "description": {
                "type": "string"
              },
              "monthly_cost": {
                "type": "string",
                "description": "A free-text cost description, e.g. 'No monthly fee' or 'Custom pricing' -- not always a plain number."
              },
              "transaction_rate": {
                "type": "string",
                "description": "Present on only some plans (e.g. one of two) -- not guaranteed present on every plan entry."
              }
            },
            "additionalProperties": true
          },
          "description": "Present only when the agent's default (no `schema` supplied) extraction returns pricing-plan-shaped data."
        }
      },
      "description": "The extracted data. Its shape follows the input `schema` when one is supplied; without `schema`, the agent returns its own default shape (e.g. a `plans` array) -- a caller supplying `schema` should expect a differently-shaped `data` object entirely.",
      "additionalProperties": true
    },
    "model": {
      "type": "string",
      "description": "The underlying extraction model the agent used, e.g. 'spark-2'."
    },
    "status": {
      "type": "string",
      "description": "'completed' on the calls observed; other values are undocumented here."
    },
    "message": {
      "type": "string",
      "description": "A plain-language summary of what the agent found and how it answered the prompt."
    },
    "success": {
      "type": "boolean",
      "description": "Whether the agent completed its task; true on the calls observed."
    }
  },
  "description": "The agent's result. A genuinely not-found run returns `{}` -- no field below is guaranteed present.",
  "additionalProperties": true
}