Skip to content
MCP server

Faultline

Check the key factual claims in AI output against live web sources

Install
Pricing
Freemium
Platforms
API, CLI, Web
Runs
Remotely
Version
0.12.1
Listing updated
3 days ago

About Faultline

Faultline is a freemium MCP server for ai models and llm tooling. It runs as a hosted remote server, so there is nothing to install locally.

Faultline splits AI-written text into claims, picks the most important factual ones, and checks each against sources it finds on the live web. A normal scan checks up to 8 claims and returns a risk band plus a list of anything it could not verify. A deep scan covers every factual claim. It can fail a CI build on the result.

The MCP server exposes a single tool, verify_claims, and needs an API key in an x-api-key header. A claim is billed only when Faultline returns a verdict backed by a source. Unverified claims and provider errors are free. The README labels the project alpha and reports 0.755 balanced accuracy on 200 human-labelled Factcheck-Bench claims.

Scans run on Gemini against sources found with Parallel, with multi-model consensus using OpenAI and Anthropic reserved for Enterprise. Hosted scans use Faultline's own provider keys. The API also produces governance records such as a risk register export and human sign-off records.

Install

Run in a terminal

claude mcp add --transport http faultline https://faultline-api.fly.dev/mcp

Add to Settings → Connectors → Add custom connector

https://faultline-api.fly.dev/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "faultline": {
      "url": "https://faultline-api.fly.dev/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "faultline": {
      "type": "http",
      "url": "https://faultline-api.fly.dev/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "faultline": {
      "serverUrl": "https://faultline-api.fly.dev/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "faultline": {
      "type": "http",
      "url": "https://faultline-api.fly.dev/mcp"
    }
  }
}

Tools this server provides (1)

Taken from the project documentation; the running server may differ.

verify_claims
Extracts claims from text and checks them against live sources.

What it does

Claim extraction
Breaks AI output into atomic factual claims.
Live-source checking
Verifies the most important claims against current web sources.
Risk band
Returns a risk level and a list of claims it could not verify.
Deep scan
Checks every factual claim instead of the top few.
CI gating
Can fail a build based on the result.
Governance records
Risk register, audit log manifest and human sign-off via the API.

Good for

  • Catch fabricated citations before publishing
  • Fact-check an agent's report in a pipeline
  • Gate pull requests that include AI-written docs
  • Produce an audit trail for reviewed AI output
  • Verify claims in a draft with a quick scan

Pricing

Freemium, from $10 a month

Free 5 scans/month; Personal $19; Pro $49; API key $10/mo plus $0.10 per billed claim

Prices change. Check faultline.nxtg.ai before you buy.

Faultline FAQ

What does a scan cost?

Free gives 5 scans a month, Personal is $19 for 100, and Pro is $49 for 500. The API key is $10 a month with 100 billed claims, then $0.10 each.

When is a claim billed?

Only when Faultline returns a verdict backed by a source it found.

How accurate is it?

The README reports 0.755 balanced accuracy on 200 human-labelled claims, and the project is in alpha.

Does it store my text?

The engine does not write scanned text to storage; signed-in scans are saved to your own history for 2 years.

Is Faultline free?

Faultline has a free plan, with paid plans for more usage or features. Paid plans start at $10 a month.

How do I add Faultline to Claude Code?

Run this command in your terminal: claude mcp add --transport http faultline https://faultline-api.fly.dev/mcp

What are the best alternatives to Faultline?

The closest alternatives to Faultline are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Faultline

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools