CoreFreeFirst-Party Ground TruthActive in MCP

Export Dataset

export_dataset(uri: string)

Fetch a full dataset by mcpseo:// uri.

Returns complete, unpaginated raw dataset tables persisted in R2/D1 storage.

Input Parameters

1 argument
ParameterTypeRequirementDefault / ConstraintsDescription
uristringrequired
Valid mcpseo:// resource URI
A dataset URI returned in a prior tool response (e.g. mcpseo://gsc/performance/example.com) for results truncated in summary envelopes.

Return Envelope (OFE / 1.0)

Open Fact Envelope

Every response adheres to the strict ofe/1.0 envelope schema, returning verified data, typed facts, entity references, and follow-up tool suggestions:

data Payload

Returns complete, unpaginated raw dataset tables persisted in R2/D1 storage.

coverage & Freshness

Reports returned count vs total items, observation timestamp as_of, and scope notes. Evaluated live per request.

resources (Dataset Exports)

If a query yields high row counts (e.g. >1,000 queries in Search Console), full unpaginated tables are persisted to R2 and linked as an mcpseo:// URI for follow-up retrieval via export_dataset.

next_actions

Provides suggested follow-up tool calls with pre-filled arguments so your AI agent can navigate from discovery to detailed inspection autonomously.

Request & Response Examples

Live MCP Payloads

Exact JSON schemas transmitted over Model Context Protocol (stdio or HTTP SSE transport):

1. Client Tool Invocation

export_dataset request.json
{
  "tool": "export_dataset",
  "arguments": {
    "uri": "mcpseo://gsc/performance/example.com"
  }
}

2. Server Envelope Response

export_dataset response.json
{
  "schema_version": "ofe/1.0",
  "domain": "core",
  "data": {
    "uri": "mcpseo://gsc/performance/example.com",
    "dataset": [
      {
        "query": "open source mcp",
        "clicks": 1420,
        "impressions": 18500,
        "ctr": 0.0768,
        "position": 2.4
      },
      {
        "query": "claude code seo",
        "clicks": 890,
        "impressions": 12200,
        "ctr": 0.0729,
        "position": 3.1
      },
      {
        "query": "headless search console api",
        "clicks": 430,
        "impressions": 5600,
        "ctr": 0.0767,
        "position": 1.8
      }
    ]
  },
  "facts": [],
  "entities": [],
  "coverage": {
    "returned": 3,
    "total": 3,
    "as_of": null,
    "scope_note": "full, untruncated dataset"
  },
  "deltas": [],
  "resources": [],
  "next_actions": []
}

LLM Agent Workflow

Claude & Cursor Integration
Step 1

Prompt Trigger

"Fetch the full unpaginated query report from the previous export link."
Step 2

Agent Decision & Reasoning

When get_search_performance or ranking tools return a resource ref because results exceed row limits, agents call export_dataset to download the full row set.

Provenance Guarantees & Error Handling

Reliability
Source Classcore
Inspection Methodcore.export_dataset
OFE Calibration Score1.00 (OFE 1.0 scale)
Cache DurationNone (Live)

Every fact emitted by Vouched includes source_class, method, observed_at ISO timestamp, and a published confidence score. LLM agents can inspect these citations to distinguish first-party verified facts (e.g. Search Console) from modeled competitor estimates.

Billing & API Keys

Free on every plan: it reads your own Google data, so there's no per-call cost. Works the same when self-hosted.

Error Conditions

  • NotFoundError: invalid or expired dataset URI

More Core Tools

2 sibling tools
Self-host on GitHub (MIT)Try on Cloud plan