Lenz Fact-Check

Fact-check factual claims against independent sources.

CommunityOnlineNo sign-inGlobalFreeRead-only

What it can do

  • Assess Claim: Fast fact-check: a verdict in ~5-10s via a 3-model panel. Use this as the default for checking a factual claim. Returns one verdict per atomic claim found in ``claim`` (True / Mostly Tru
  • Verify Claim: Start a deep, multi-step fact-check (research → debate → adjudication). Reserve for high-stakes claims that warrant a thorough, sourced investigation — it takes ~90s and uses scarce deep
  • Select Claims: Resolve a `needs_input` verification by choosing which claim(s) to run. When `get_verification` returns ``status: needs_input`` with reason ``multi_claim`` or ``clarification_required``

What data it sees

Do you need an account

No: the server works without sign-in

Fact-check factual claims against independent sources. assess gives a fast multi-model verdict (True → False) with bucketed confidence in ~5–10s; verify runs a deeper research → debate → panel investigation for high-stakes claims and returns sourced, pollable results. ask answers grounded follow-ups on a completed verification. Verdicts are directional, not absolute. Connect via OAuth or a free Lenz API key.

How to connect

How to connect Lenz Fact-Check to Claude

The server address to paste into Claude settings:

https://fact-check--lenz.run.tools

  1. Open Claude (claude.ai or the desktop app).
  2. Go to Settings → Connectors.
  3. Click Add custom connector.
  4. Paste the server address copied below into Remote MCP server URL and click Add.
  5. In a chat, click + → Connectors and switch the new connector on.

Custom connectors are available on Free, Pro, Max, Team and Enterprise plans (Free is limited to one). On Team and Enterprise an organization Owner adds the connector first under Organization settings → Connectors.

Authorization

After you click Add, a sign-in window for the service opens. Sign in with your own account and approve access. Claude never sees your password.

How to connect Lenz Fact-Check to ChatGPT

The server address to paste into ChatGPT settings:

https://fact-check--lenz.run.tools

  1. Open ChatGPT in a browser (chatgpt.com). A Plus, Pro, Business, Enterprise or Edu plan is required.
  2. Turn on developer mode once: Settings → Apps → Advanced settings → Developer mode.
  3. Open Settings → Connectors and click Create.
  4. Fill in the form: Name (anything), Description (one line about what the service does), MCP server URL (copy it below).
  5. Under Authentication choose OAuth if the service requires sign-in, otherwise None. Click Create.
  6. In a new chat open + → Apps/Connectors and enable the connector.

OpenAI has renamed this section before (Connectors → Apps/Plugins). If the label differs, search settings for "developer mode". On Business/Enterprise workspaces an admin must allow custom connectors first.

Authorization

On first use ChatGPT opens the service's sign-in window. Sign in and approve access.

How to connect Lenz Fact-Check to Cursor

The server address to paste into Cursor settings:

https://fact-check--lenz.run.tools

Fastest: click Open in Cursor below and confirm the prompt.

Manually:

  1. In Cursor open Settings → Cursor Settings → MCP and click Add new global MCP server.
  2. Paste the JSON copied below into ~/.cursor/mcp.json (per project: .cursor/mcp.json in the repo root).
  3. Save the file. The server appears in the MCP list; authorize it there if asked.

Authorization

If the service needs sign-in, an authorize button appears next to the server in the MCP list.

Server tool list (6)

Raw names from tools/list. Only developers need these.

assess_claimFast fact-check: a verdict in ~5-10s via a 3-model panel. Use this as the default for checking a factual claim. Returns one verdict per atomic claim found in ``claim`` (True / Mostly True / Mixed / Mostly False / False) plus a bucketed confidence. Cheaper and faster than `verify_claim`; no detailed sources. If the text is ambiguous it returns candidate readings to re-assess; if it contains no checkable claim it says so. ``language`` is an optional ISO code for the response language.
verify_claimStart a deep, multi-step fact-check (research → debate → adjudication). Reserve for high-stakes claims that warrant a thorough, sourced investigation — it takes ~90s and uses scarce deep-check quota. Returns a ``task_id`` immediately; call `get_verification(task_id)` until its status is ``completed`` to retrieve the sourced result. For a quick answer use `assess_claim` instead.
select_claimsResolve a `needs_input` verification by choosing which claim(s) to run. When `get_verification` returns ``status: needs_input`` with reason ``multi_claim`` or ``clarification_required``, call this with that ``task_id`` and a ``claims`` list containing one or more of the offered claim texts (from the `claims`/`candidates` it returned). Each selected claim starts its own deep verification; poll each returned ``task_id`` with `get_verification`. (Not for ``duplicate_found`` — use the surfaced `similar_claims` there instead.)
get_verificationRetrieve (or poll) a deep `verify_claim` result by its ``task_id``. While running, returns ``status: processing``. If the claim needs a decision (multiple claims, ambiguity, or a near-duplicate), returns ``status: needs_input`` with the options. When finished, returns ``status: completed`` with the verdict, confidence, executive summary, top sources, and the ``verification_id`` — pass that to `ask_followup` for a grounded follow-up. (No Lenz link: deep-check claims are private to the caller, so a claim-page link would be share-gated rather than publicly viewable.)
check_usageReturn remaining Lenz quota for the calling key, per capability. Reports remaining ``assess`` and ``verify`` capacity (and the current plan). Call this before a batch of checks to avoid hitting a quota wall mid-run.
ask_followupAsk a grounded follow-up question about a completed `verify_claim` result. The `verify_claim`/`get_verification` result is trimmed; this reads the FULL evidence (research, debate, per-panelist adjudication, all sources) the trimmed payload omits and answers ``question`` from it. Pass the ``verification_id`` returned by `get_verification` (NOT a task_id). Only works on a completed `verify_claim` — not on `assess_claim` results. Uses your `ask_followup` quota. The conversation is kept server-side per verification, so ask follow-ups sequentially rather than in parallel. ``language`` is an optional ISO code for the reply language.