How to Use the Hallucination Detector Prover MCP in Windsurf

Q: How does Hallucination Detector Prover stop Windsurf from making up API endpoints?

It intercepts Cascade's plans using the validatehallucinationgrounding tool to verify every referenced endpoint against real documentation. If the agent cannot provide a specific URL or version cutoff, the step is flagged as ungrounded. This keeps your MCP runs bounded by actual API limits.

Q: What happens when a claim fails verification in Cascade?

The validatehallucinationgrounding tool rejects the assertion and returns a failure status. This forces Cascade to revise its plan, look for better sources, or explicitly state what it does not know. Your agent is kept honest by refusing to let unverified assumptions pass.

Q: How does this MCP Server handle the privacy of my Windsurf workspace data?

All factual claims, citations, and text inputs sent to the validatehallucinationgrounding tool are processed inside Vinkius's isolated sandbox. No workspace data is saved or used for model training. The environment is destroyed immediately after validation, keeping your intellectual property safe.

Force Windsurf to verify its own claims autonomously before Cascade writes a single line of ungrounded code or documentation.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Hallucination Detector Prover MCP to Windsurf

Create your Vinkius account to connect Hallucination Detector Prover to Windsurf and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup Hallucination Detector Prover with Windsurf

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Grounding Cascade agent steps with verified sources

The `validate_hallucination_grounding` tool acts as an automated epistemic gatekeeper within your Windsurf Cascade workspace connected to this MCP Server. When Cascade plans a multi-step execution, this tool forces the agent to cite exact authors, dates, and DOIs for its technical assertions before proceeding. This prevents Cascade from fabricating library features or API parameters during autonomous runs. You get verified, traceably accurate code generation because the agent must prove its knowledge boundaries before writing files.

Calibrating confidence in Windsurf multi-tool chains

The `validate_hallucination_grounding` tool evaluates the certainty of every technical claim your Windsurf agent proposes during a complex plan. It assigns distinct confidence levels, making Cascade distinguish between a verified official documentation page and a random forum post. By forcing this calibration, Windsurf stops executing tasks based on shaky assumptions. You avoid debugging code built on outdated blog posts because the agent explicitly flags unverified claims.

Eliminating internal contradictions in Windsurf plans

This MCP Server runs deep cross-reference checks across the entire context window of your Cascade session. The `validate_hallucination_grounding` tool scans for conflicting statements, like claiming a port is open in one step and closed in another. When Cascade attempts to chain multiple terminal commands, the tool intercepts self-contradictory logic. It forces the agent to reconcile these issues before executing a broken script.

Setup guide

Set up Hallucination Detector Prover MCP in Windsurf

Prerequisites

Windsurf IDE installed (macOS, Windows, or Linux)
Active Vinkius subscription with a valid endpoint token

1

Open MCP configuration

Click the Cascade assistant icon in the sidebar, then click the hammer icon (🔨) at the top of the panel. Select "Configure" to open ~/.codeium/windsurf/mcp_config.json.
2

Add the Hallucination Detector Prover MCP

Paste the JSON snippet shown on the right into the mcpServers object. Replace [YOUR_TOKEN_HERE] with your endpoint token from cloud.vinkius.com.
3

Refresh MCPs

Go back to the hammer icon (🔨) in Cascade and click "Refresh". Windsurf will detect the new server. No full restart is needed — the connection is hot-reloaded.
4

Verify in Cascade

Start a new Cascade conversation and ask something like "Show my Hallucination Detector Prover payment history." If connected, Cascade will call the Hallucination Detector Prover tools directly. You will see a green dot next to the server name in the MCP panel.

mcp_config.json

{
  "mcpServers": {
    "hallucination-detector-prover-mcp": {
      "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    }
  }
}

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Hallucination Detector Prover. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Hallucination Detector Prover now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Hallucination Detector Prover MCP in Windsurf

It intercepts Cascade's plans using the `validate_hallucination_grounding` tool to verify every referenced endpoint against real documentation. If the agent cannot provide a specific URL or version cutoff, the step is flagged as ungrounded. This keeps your MCP runs bounded by actual API limits.

Yes, this MCP Server inspects the text files and context Cascade gathers during its workspace analysis. The `validate_hallucination_grounding` tool checks these files to separate objective facts from developer opinions. You receive clear warnings when the agent treats a subjective comment in your codebase as an absolute system requirement.

The verification step adds a minor validation delay, but it prevents hours of debugging broken code. The `validate_hallucination_grounding` tool runs inline as Cascade builds its plan, catching fabrications before execution. This trade-off ensures your autonomous agent does not run off the rails.

The `validate_hallucination_grounding` tool rejects the assertion and returns a failure status. This forces Cascade to revise its plan, look for better sources, or explicitly state what it does not know. Your agent is kept honest by refusing to let unverified assumptions pass.

All factual claims, citations, and text inputs sent to the `validate_hallucination_grounding` tool are processed inside Vinkius's isolated sandbox. No workspace data is saved or used for model training. The environment is destroyed immediately after validation, keeping your intellectual property safe.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript