Browser Proof
A verification component for agents that checks claims on public webpages and returns structured results with evidence text, screenshots, and deterministic JSON.
README
Browser Proof
Browser Proof is evidence-backed web verification for agents.
Browser Proof is a verification component for agents that checks claims on public webpages and returns structured results with evidence text, screenshots, and deterministic JSON.
It is not a general-purpose browser automation product.
It is designed for situations where an agent needs external confirmation with evidence.
What it does
Browser Proof helps agents:
- verify whether a claim appears on a public webpage
- extract supporting evidence text
- capture a screenshot for auditability
- return a stable JSON response for downstream systems
Production access
Production Browser Proof access now requires an API key.
- MCP remote access requires
X-API-Key - HTTP API access requires
X-API-Key - Registry visibility remains public for discovery, but live usage is not anonymous
Current launch plans:
trial: 20 credits / monthstarter: 1000 credits / monthpro: 10000 credits / month
Initial credit costs:
verify_claim: 5 creditslist_runs: 1 creditget_run: 1 creditget_run_screenshot: 1 credit
If you need a key for evaluation or paid use, contact RePrompt through the support or sales channel listed for this project.
Typical use cases
- Pricing check: confirm whether a price is actually shown on a page
- Policy validation: verify whether required language appears on a landing page
- Listing QA: confirm whether a job post, real-estate listing, or product page still matches expected facts
- Citation-backed research support: retrieve the exact text used as evidence
Quick start
HTTP API
curl -X POST https://proof.reprompt.jp/api/browser-proof \
-H "X-API-Key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"url": "https://example.com",
"task": "Check whether this page says it is for use in documentation."
}'
Example response:
{
"success": true,
"answer": "Yes, the page states that it is for use in documentation examples.",
"evidence_text": "This domain is for use in documentation examples without needing permission.",
"source_url": "https://example.com",
"checked_at": "2026-04-08T15:00:00Z",
"screenshot_url": "supabase://browser-proof-evidence/example.com/root/2026-04-08T15-00-00-000Z.jpg",
"error": null
}
Notes:
screenshot_urlis currently returned as an internal storage reference string.- Clients that need a retrievable screenshot URL should use the run retrieval interfaces or MCP tools.
MCP connection
Browser Proof exposes remote MCP endpoints.
- Recommended (Streamable HTTP):
https://proof.reprompt.jp/mcp - SSE:
https://proof.reprompt.jp/mcp/sse
See:
server.jsonexamples/docs/mcp.md
HTTP endpoints
POST /api/browser-proofGET /api/runsGET /api/runs/:idGET /api/runs/:id/screenshotGET /health
Authentication
Production use requires an API key.
- Preferred header:
X-API-Key: YOUR_API_KEY - Bearer-style client configuration may be supported for compatibility, but
X-API-Keyis the canonical format - Do not store live keys in this repository
- Keys are issued individually and may be rate-limited or credit-limited by plan
See:
docs/auth.mdexamples/mcp/
Limits, retention, and privacy
See:
docs/limits.mddocs/retention.mddocs/privacy.md
If a value is not finalized, mark it as planned or to be finalized. Do not present tentative values as final.
Repository purpose
This repository is the public shelf for Browser Proof:
- discovery
- connection details
- interface reference
- examples
- integration guidance
Implementation details live separately in the private core repository.
Related files
agent-card.jsonserver.jsonopenapi.yamlschema/examples/docs/
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.