consistency-gate

Check whether a claim contradicts a known set of facts.

CommunityOnlineNo sign-inGlobalFreeCan modify data

What it can do

  • Check Claim: NL-gate (kernen): udtræk SPO-triple(r) fra en naturlig-sprogs-PÅSTAND (Sonnet via LiteLLM, R1) og tjek hver mod det grundede SPO-fakta-lager (i PUBLIC_DEMO-mode: det OFFENTLIGE demo-fakta
  • Check Triple: Rå SPO-gate, deterministisk (ingen LLM — billig, testbar kerne, ÅBEN+GRATIS i PUBLIC_DEMO-mode, ingen rate-grænse §2). Konflikt = samme normaliseret subjekt+prædikat, andet normaliseret
  • List Facts: Alle fakta hvor `entity` er subjekt ELLER objekt (transparens/audit). `entity` normaliseres FØR match — case/whitespace-varianter matcher samme række. Returtypen er ALTID en liste (§8-adju

What data it sees

Do you need an account

No: the server works without sign-in

Check whether a claim contradicts a known set of facts. Extract subject-predicate-object triples from a natural-language statement (or pass a raw triple) and check each against a grounded fact store. Returns an auditable receipt with a verdict of consistent, contradicts, or unverifiable, plus every conflict it found, never a bare true/false. Advisory and read-only: it flags consistency, it never blocks, decides, or writes for you, and checking a claim never stores it. Includes a deterministic no-LLM triple check for cheap, testable runs, plus a fact-listing tool for transparency. Public demo runs against an open demo fact store, not your data.

How to connect

How to connect consistency-gate to Claude

The server address to paste into Claude settings:

https://consistency-gate--humanified.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 consistency-gate to ChatGPT

The server address to paste into ChatGPT settings:

https://consistency-gate--humanified.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 consistency-gate to Cursor

The server address to paste into Cursor settings:

https://consistency-gate--humanified.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 (3)

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

check_claimNL-gate (kernen): udtræk SPO-triple(r) fra en naturlig-sprogs-PÅSTAND (Sonnet via LiteLLM, R1) og tjek hver mod det grundede SPO-fakta-lager (i PUBLIC_DEMO-mode: det OFFENTLIGE demo-fakta-lager, IKKE dine data — se scope_note). Returnerer en auditerbar consistency-receipt: {claim, extracted_triples[], conflicts[], verdict, checked_against_n_facts, scope_note}. `verdict` ∈ {consistent, contradicts, unverifiable} — ALDRIG true/false. ADVISORY: flager konsistens, blokerer/beslutter ALDRIG for dig (never-write-for-you). Persisterer ALDRIG selve påstanden (check ≠ gem — kun add_fact skriver til lageret, og add_fact er FRAVÆRENDE i PUBLIC_DEMO-mode). PUBLIC_DEMO-mode: rate-limitet (default 5/min + 30/time, ÉN global demo-bucket, §2/§8 BLOCKER 3). Over grænsen returneres `{error:"rate_limited", retry_after_s, scope_note}` — ALDRIG et consistent-formet svar. Bogholderi-fejl i rate-limiteren -> deny (fail-safe).
check_tripleRå SPO-gate, deterministisk (ingen LLM — billig, testbar kerne, ÅBEN+GRATIS i PUBLIC_DEMO-mode, ingen rate-grænse §2). Konflikt = samme normaliseret subjekt+prædikat, andet normaliseret objekt. Returnerer ALLE konflikter (ikke kun første). `verdict` ∈ {consistent, contradicts} — ALDRIG true/false. Read-only mod lageret (i PUBLIC_DEMO-mode: det offentlige demo-fakta-lager).
list_factsAlle fakta hvor `entity` er subjekt ELLER objekt (transparens/audit). `entity` normaliseres FØR match — case/whitespace-varianter matcher samme række. Returtypen er ALTID en liste (§8-adjudikering: ingen dict-wrap/scope_note-injektion her, det ville være en regression mod default-modens kontrakt). PUBLIC_DEMO-mode: listen viser udelukkende rækker fra det OFFENTLIGE demo-fakta-lager (demo-facts.json) — ikke fleet/TIER-1/cortex-data (§3 PUBLIC-SAFE).