Verification Layer For AI Agents

Stop your AI from being wrong.

Ground Truth lets AI agents verify claims, check live data, compare competitors, inspect APIs, and validate assumptions before acting.

View Pricing See MCP Setup

Live data checks Direct API or MCP Cloudflare Workers

What Ground Truth verifies

Ground Truth helps agents check the kind of facts that are most likely to drift, break, or get invented under pressure.

Verify a pricing claim

Pull the live pricing page before your agent quotes a number like "Notion costs $8 per user per month."

check_pricing

Check whether a competitor exists

Search npm or PyPI before your agent says there is no alternative in a category.

estimate_market

Validate an API endpoint

Confirm the URL exists, responds, and looks real before recommending it to a user or team.

check_endpoint

Compare package popularity

Compare live package metadata instead of guessing which package is more active or widely used.

compare_competitors

Test a market assumption

Turn assumptions like "this category is still small" into pass/fail checks against live data.

test_hypothesis

Confirm whether a support policy applies

Check public help, support, or policy pages before your agent repeats a claim as fact.

verify_claim

Why AI agents need verification

Training data does not tell you what changed this week. Ground Truth gives agents a last-mile check before they answer, recommend, or act.

Pricing changes quietly

Agents often repeat old prices long after a plan page changed. Live verification catches that drift.

Endpoints look real even when they are not

A confident recommendation is not the same as a working API. Ground Truth checks the endpoint first.

Competitive claims get invented

Agents say "no competitors" or "most popular" without checking the current market. Registry data gives them a way to prove it.

Policies and support terms move

Support plans, public terms, and help-center language change over time. Verification keeps answers tied to the live source.

Example workflows

Use Ground Truth when an answer should be checked before it leaves the model.

Pricing claim

Before an agent says "Stripe has a free tier," it checks the live pricing page and returns what it found.

Competitor existence

Before an agent says there is no alternative to Prisma for edge deployments, it searches the registry for real packages.

API validation

Before an agent recommends an endpoint in docs or support, it confirms the endpoint responds.

Package comparison

Before an agent says Vue has overtaken React, it compares live package metadata side by side.

Market assumption

Before an agent says the MCP ecosystem is still tiny, it runs a count-based hypothesis test.

Support policy confirmation

Before an agent repeats a support entitlement, it checks the current public support page for evidence.

Free vs Pro

Start with a quick check. Upgrade when your agents need broader verification coverage and higher limits.

Free

Quick validation

Best for basic endpoint checks and limited verification needs.

Only check_endpoint
100 requests per calendar month
No API key required for the free check

Pro

Full verification layer

$9/month

Built for teams that want Ground Truth wired into an always-on AI workflow.

Requires X-API-Key
5,000 requests per calendar month by default
Usage tracked per API key and tool
Competitor comparison
Claim verification
Market checks
Structured reports
Priority response

Get Pro Access Try The Free Check

API examples

Use Ground Truth directly over HTTP if you want verification in a script, backend, or agent loop.

Direct API `curl` call

Initialize the MCP session, then verify a pricing claim against a live pricing page.

SESSION_ID="$(curl -i -s -X POST https://ground-truth-mcp.anishdasmail.workers.dev/mcp \
  -H "Accept: application/json, text/event-stream" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","method":"initialize","params":{"protocolVersion":"2025-03-26","capabilities":{},"clientInfo":{"name":"ground-truth-example","version":"1.0.0"}},"id":0}' | tr -d '\r' | awk '/^mcp-session-id:/ {print $2}')"

curl -X POST https://ground-truth-mcp.anishdasmail.workers.dev/mcp \
  -H "Accept: application/json, text/event-stream" \
  -H "Content-Type: application/json" \
  -H "Mcp-Session-Id: $SESSION_ID" \
  -H "X-API-Key: $GROUND_TRUTH_API_KEY" \
  -d '{
    "jsonrpc": "2.0",
    "method": "tools/call",
    "params": {
      "name": "check_pricing",
      "arguments": {
        "url": "https://stripe.com/pricing"
      }
    },
    "id": 1
  }'

JavaScript `fetch` example

Compare package popularity from code.

const initResponse = await fetch("https://ground-truth-mcp.anishdasmail.workers.dev/mcp", {
  method: "POST",
  headers: {
    "Accept": "application/json, text/event-stream",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    jsonrpc: "2.0",
    method: "initialize",
    params: {
      protocolVersion: "2025-03-26",
      capabilities: {},
      clientInfo: {
        name: "ground-truth-example",
        version: "1.0.0",
      },
    },
    id: 0,
  }),
});

const sessionId = initResponse.headers.get("mcp-session-id");

if (!sessionId) {
  throw new Error("Missing mcp-session-id from initialize response");
}

const response = await fetch("https://ground-truth-mcp.anishdasmail.workers.dev/mcp", {
  method: "POST",
  headers: {
    "Accept": "application/json, text/event-stream",
    "Content-Type": "application/json",
    "Mcp-Session-Id": sessionId,
    "X-API-Key": process.env.GROUND_TRUTH_API_KEY,
  },
  body: JSON.stringify({
    jsonrpc: "2.0",
    method: "tools/call",
    params: {
      name: "compare_competitors",
      arguments: {
        packages: ["react", "vue"],
        registry: "npm",
      },
    },
    id: 1,
  }),
});

const result = await response.json();
console.log(result);

MCP setup

MCP stands for Model Context Protocol. If you use Claude Desktop or Cursor, Ground Truth can plug in as a verification tool so your agent checks live data before it responds.

Claude Desktop

Add Ground Truth to ~/Library/Application Support/Claude/claude_desktop_config.json.

{
  "mcpServers": {
    "ground-truth": {
      "url": "https://ground-truth-mcp.anishdasmail.workers.dev/mcp",
      "headers": {
        "X-API-Key": "gt_live_your_key_here"
      }
    }
  }
}

Cursor

Add Ground Truth to .cursor/mcp.json in your project or ~/.cursor/mcp.json globally.

{
  "mcpServers": {
    "ground-truth": {
      "url": "https://ground-truth-mcp.anishdasmail.workers.dev/mcp",
      "headers": {
        "X-API-Key": "gt_live_your_key_here"
      }
    }
  }
}

Use cases

Ground Truth fits anywhere an AI agent needs a final check before sharing an answer or taking action.

Support

Verify pricing claims, confirm whether a support policy applies, and check whether an API endpoint a customer asks about actually exists.

Product

Test market assumptions, check whether a competitor exists, and compare package popularity before you lock in a direction.

Legal

Confirm support and policy language on live public pages before repeating it internally or externally.

Market research

Compare competitor pricing, count category players, and turn broad assumptions into structured live checks.