Deckrd is three capabilities: Detect AI in chats, calls, and video;
Protect your voice from cloning;
Verify a voice is genuine for people you trust.
@deckrd/sdk
ships Detect · text today. Scoring runs locally; message content never reaches our servers.
The npm SDK exposes Detect for integrators today. Protect and Verify APIs follow once those layers leave prototype.
Eleven local detectors score conversations in real time. High score = likely AI. Runs entirely in your runtime.
AntiClone makes your voice unusable to AI cloners while sounding natural to people.
Verify the live human, cryptographically. Peer to peer for people, organisation-managed for enterprise. Nothing about the voice leaves the device; no voiceprint is ever stored. API follows the prototype. Learn about Verify →
Runs in your Node, browser, edge, or RN runtime. We never see your users' messages.
Linguistic style, glyphs, response timing, mirror, persona consistency, and more. When generators improve at one signal, the others still hold. Free tier ships with the npm package; paid tiers get fresh tunings via OTA every week.
npm install @deckrd/sdk + a score() call. Drop-in for Next.js, Cloudflare Workers, Vercel Edge, Deno, React Native, Bun, or any modern Node. No build pipeline, no external services, no model hosting.
Free tier works without a license key. Add one when you're ready to remove the attribution badge or upgrade volume.
Each score() is one billable detection. You decide when to re-evaluate
(every message, every five, or on-demand). There is no implicit cost
for buffering.
The SDK makes exactly three network calls: licence validation, usage metering and detector updates. None carries message content. An independent security audit of Voice Proof and the SDK is planned for the next release cycle.
| Call | Carries | When | Content? |
|---|---|---|---|
| Licence validation | SHA-256 of the key | Lazy, cached 1h | never |
| Usage metering | Counts + label totals | Batched, periodic | never |
| Detector updates | One-way pull of lexicons | Weekly + manual refresh | never |
Detection runs entirely on your runtime. We can't see your users'
messages even if we wanted to. The 11 text detectors live in
@deckrd/detection-core;
they're heuristic functions over message arrays, no remote inference,
no model hosting.
One detection = one call to score(). You control granularity (every message vs. every fifth vs. on-demand) and predictability follows.
Organisation-managed Verify: directory, policy and audit trail run in your environment.
Talk to usRequest an API key. Free tier: 10,000 detections a month.