Agent Conversion — AI Website Conversion Testing
MCP server that runs browser audits to test whether AI agents can complete website
Answered our latest tool-list check 3 hours ago with 2 tools. It will be marked verified after 3 successful checks in 7 days.
- Platforms
- API
- Runs
- Remotely
- Version
- 0.1.1
- Listing updated
- today
Similar MCP servers
About Agent Conversion — AI Website Conversion Testing
Agent Conversion — AI Website Conversion Testing is an MCP server for browser automation. It runs as a hosted remote server, so there is nothing to install locally.
The server lets an MCP-capable agent send a public website and a plain-language commercial goal, such as finding a plan's monthly price, starting a self-service signup, or reaching a demo request form. A real browser journey is then run against the site, and the result is returned as an observed outcome with supporting evidence steps rather than a single readiness score.
The workflow has two tools. One starts an audit asynchronously and returns an audit ID, a report URL and a suggested retry interval; the other polls for the report, which includes whether the task was completed, the evidence trail, a summary, recommendations and a score when the run supports one. The site documentation says audits may take a few minutes and that the agent does not buy, pay or submit real personal data.
It suits teams checking how well their pricing, signup or sales pages work for AI browsing agents, and developers who want an agent to report on such journeys. The endpoint needs no API key, but audit starts are rate-limited and the service is described as an early public capability.
Install
Run in a terminal
claude mcp add --transport http agent-conversion https://agent-conversion-coral.vercel.app/api/mcpAdd to Settings → Connectors → Add custom connector
https://agent-conversion-coral.vercel.app/api/mcpAdd to ~/.cursor/mcp.json
{
"mcpServers": {
"agent-conversion": {
"url": "https://agent-conversion-coral.vercel.app/api/mcp"
}
}
}Add to .vscode/mcp.json
{
"servers": {
"agent-conversion": {
"type": "http",
"url": "https://agent-conversion-coral.vercel.app/api/mcp"
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"agent-conversion": {
"serverUrl": "https://agent-conversion-coral.vercel.app/api/mcp"
}
}
}Add to Grok Bot desktop app → Connect apps → Add a custom connector
https://agent-conversion-coral.vercel.app/api/mcpEnter a Name (for example "Agent Conversion — AI Website Conversion Testing"), choose Type: Remote HTTPS, paste this as the Server URL and choose Add plugin.
Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"agent-conversion": {
"type": "http",
"url": "https://agent-conversion-coral.vercel.app/api/mcp"
}
}
}Using several servers? Build one config file with the MCP config generator.
Tools this server provides (2)
Read directly from the server.
1 of 2 tools says it's read-only. 1 of 2 tools says it can delete or change data.
From the server's own tool annotations; not verified by us.
- test_conversionSays it can change data
- Start a real Agent Conversion audit: an autonomous browsing agent opens the given website and tries to complete the task, then the run is scored 0-100 with problems and recommendations. Use this when someone wants to know whether an AI agent can find something or complete a commercial step on a specific website, e.g. "Can an agent find the Pro plan's monthly price on example.com?" or "Can an agent
- get_conversion_reportSays read-only
- Get the status and, once finished, the result of an audit started with test_conversion. Pass the audit_id returned by test_conversion (a report URL also works). status is "pending" while the agent is still running: wait retry_after_seconds and call again. If stale is true, the audit will never finish; start a new one. When status is "completed", result contains: task_completed (did the agent reach
What it does
- Browser-based journey testing
- Runs the requested task in a real browser against a public website instead of a static check.
- Evidence-backed outcomes
- Reports the observed result along with the steps taken as evidence.
- Asynchronous audit flow
- Starts an audit and returns an ID so the client can poll for the finished report.
- Separated failure reasons
- Distinguishes site refusals from incomplete or infrastructure-failed runs, where the evidence supports it.
- No API key required
- The remote MCP endpoint connects over streamable HTTP without credentials, though audit starts are rate-limited.
Good for
- Checking whether an AI agent can find a SaaS plan's monthly price
- Testing whether a self-service signup flow can be completed by an agent
- Verifying that a demo or contact-sales path is reachable by automated browsing
- Reviewing evidence steps and recommendations after a conversion test
Agent Conversion — AI Website Conversion Testing FAQ
Does the server need an API key to connect?
No. The site says the remote endpoint is Streamable HTTP and requires no API key, though audit starts are rate-limited.
Does starting an audit run a browser journey?
Yes, starting an audit with test_conversion launches a real browser journey. Connecting and listing tools do not start an audit.
Will the agent buy anything or submit personal data?
The site states the agent never buys, pays or submits real personal data.
How long does an audit take?
The site says audits execute real browser journeys and may take a few minutes.
How do I add Agent Conversion — AI Website Conversion Testing to Claude Code?
Run this command in your terminal: claude mcp add --transport http agent-conversion https://agent-conversion-coral.vercel.app/api/mcp
What are the best alternatives to Agent Conversion — AI Website Conversion Testing?
The closest alternatives to Agent Conversion — AI Website Conversion Testing are Browser Use, Agent Browser, Chrome DevTools MCP, Lighthouse Audit and MCP Browser Extension.
Where this comes from
- Description: written from the project's website and its MCP Registry entry.
- Status and tools: checked automatically by AIPoweredBot on 19 November 2026.
- Install steps: from the official MCP Registry entry.
