Skip to main content
Web Summit Lisbon, 9 to 12 Nov. Meet the founder in Lisbon.Book time

Software we build and operate.

Reply Intelligence Group Ltd owns and runs six products. ReplyOS, ReplyBench and ReplyScore are live products for UK dental and aesthetics practices. VetReady is a free resource for independent veterinary practices. Slipmark is open source and free to use on GitHub; Weft is in early access and not yet generally available. Both are for teams shipping AI features. None of them is required to hire us for a service.

Products for dental and aesthetics practices.

Three products for UK dental and aesthetics practices. ReplyScore measures whether AI assistants recommend a practice. ReplyBench measures how it handles the enquiries it already gets. ReplyOS handles enquiries across channels. Each is standalone, and none is required to hire us for a service.

No 01
Stage
Get found
Status
Live

ReplyScore

Find out if AI recommends your practice, before your competitors do.

When patients ask AI assistants for the best dentist near them, someone gets named. ReplyScore crawls your site, puts real patient questions to AI engines, and scores your visibility against audited UK peer practices, with prioritised, paste-ready fixes.

For
UK dental practices.
Limit
A score reflects what AI assistants said on the day it ran and can change.
Pricing
Free teaser score · Full report £49 (launch pricing)
No 02
Stage
Convert
Status
Live

ReplyBench

Find out if competitors are winning enquiries you already paid for.

ReplyBench scores your enquiry journey, follow-up readiness, and competitor positioning, then estimates where revenue may be leaking before it reaches your diary. Not an SEO audit: it measures how you handle the enquiries you already get.

  • Scores six leakage dimensions into one ReplyBench Score
  • Indicative revenue-at-risk estimate from your inputs
  • Automated recommendations and shareable report link
  • Pairs with ReplyOS: diagnose first, then fix enquiry handling at scale
For
Dental and aesthetics practices.
Limit
Revenue at risk is an estimate from the inputs given, not booked or actual revenue.
Pricing
Free Snapshot (£0) | Revenue Leakage Audit £495, launch pricing £249Audit fee credited in full against ReplyOS setup.

ReplyBench dimensionsIndicative

Illustrative scores. Run a free snapshot at replybench.co.uk for your practice.

No 03
Stage
Operate
Status
Live

ReplyOS®

Protect treatment revenue from first patient contact.

A UK-first vertical SaaS that identifies, governs and recovers dormant commercial opportunity across enquiry, booking and follow-up channels. Built dental-first, extending across aesthetics, optometry and physiotherapy.

  • Omnichannel inbound: SMS, WhatsApp, Instagram, Facebook, email and web forms
  • AI-assisted response with operator-grade controls and human handoff
  • Booking conversion flows inside conversations with full attribution
  • Revenue command centre: pipeline visibility, capture rate, enquiry-to-revenue chain
  • Plan tiers: Core, Pro and Scale: essential infrastructure to advanced reporting
For
UK dental and aesthetics practices.
Limit
Staff still need to take over where a person is needed.
ReplyOS signalExample
  • 09:14SMSinbound.received
  • 09:14Systemattribution.linked
  • 09:18Bookingslot.confirmed+£124
  • 09:18Revenuerecordedpipeline

Plan tiers

  • CoreEssential revenue infrastructure: webchat, SMS, email
  • ProAll channels plus automation and CRM sync
  • ScaleAdvanced reporting, control surfaces and governance

Labs: for teams shipping AI.

The reliability and interface tools our products are built on, available to teams shipping AI features of their own.

  • No 04Agent verification

    Slipmark

    Make AI agents prove what they checked.

    Slipmark makes AI agents prove what they checked. Every agent verdict is bound to a SHA-256 hash of the exact files the agent read. Before anything ships, the gate re-hashes them and refuses if anything changed. It runs as a GitHub Action. It also includes drift diagnosis for AI features. Open source under the Apache-2.0 licence. Slipmark Cloud, the hosted layer at app.slipmark.dev, adds a publicly anchored record of every verdict, review rules enforced as a required GitHub check, alerts and audit evidence.

    Status
    Open source
    Version
    v0.5.0
    Licence
    Apache-2.0
  • No 05Generative UI runtime

    Weft

    Adaptive interfaces within defined rules.

    Weft is a runtime where a model assembles each screen from a closed catalogue of components you define. Every screen is checked against a schema before it reaches a device, and every screen served is recorded so it can be reproduced. A surface can be pinned back to a hand-authored fallback. It is in early access and is not yet generally available.

    Status
    Early access
    Access
    Agreed per team

Products for veterinary practices.

No 06
Status
Live

VetReady

Free plain-English guides to each duty in the CMA vets Order, with the dates and the GOV.UK sources. A checker that reads a practice website and reports what is missing opens in November 2026.

Questions about the products.

What is Reply Intelligence Group?

Reply Intelligence Group Ltd (Company No. 17040048) is a Watford company, registered in England and Wales in February 2026. It is a UK AI implementation and software company that sells implementation services and owns ReplyOS, ReplyScore and ReplyBench for UK dental and aesthetics practices, VetReady for independent veterinary practices, and Slipmark and Weft for product teams. It is not the US firm at replyintelligence.com.

What is ReplyOS and who is it for?

ReplyOS is UK-first software for dental and aesthetics practices. It captures enquiries across SMS, WhatsApp, social channels, email and web forms; assists with AI-guided responses; converts conversations into bookings; and attributes revenue back to the originating message. It is built for practice owners and operations leads, not generic marketing teams.

What is Slipmark?

Slipmark makes AI agents prove what they checked. Every agent verdict is bound to a SHA-256 hash of the exact files the agent read. Before anything ships, the gate re-hashes them and refuses if anything changed. It runs as a GitHub Action and also includes drift diagnosis for AI features. Version 0.2.0 is live and open source under the Apache-2.0 licence. The code is at github.com/founder-ship-it/Slipmark.

How does Slipmark bind a verdict to what the agent read?

The agent's verdict is bound to a SHA-256 hash of the exact files it read. Before anything ships, the gate re-hashes those files. If anything changed, it refuses. Attestation is bound to an artifact rather than asserted by the agent. Slipmark's drift diagnosis lets you codify behavioural locks. In ReplyOS we use it to lock tenant isolation, enquiry capture and email threading.

What is ReplyScore?

ReplyScore is an AI visibility benchmark for UK dental practices. It crawls a practice site, puts real patient questions to AI engines, and scores visibility against audited UK peer practices, with prioritised, paste-ready fixes. A free teaser score is available at replyscore.co.uk; the full report is £49 at launch pricing.

What is ReplyBench and how does it relate to ReplyOS?

ReplyBench is a live enquiry leakage diagnostic at replybench.co.uk. It scores six dimensions (enquiry speed, CTA clarity, follow-up readiness, competitor positioning, trust signals, and revenue tracking) and combines them into a single ReplyBench Score with an indicative revenue-at-risk estimate. The free Snapshot is instant; a Revenue Leakage Audit is £495 (launch pricing £249), credited in full against ReplyOS setup. ReplyBench diagnoses where leakage occurs; ReplyOS provides the infrastructure to fix enquiry handling, follow-up and attribution at scale.

What is Weft and who is it for?

Weft is a runtime for product teams in which a model arranges each screen from a closed catalogue of the team's own components and design tokens. It is for teams that want interfaces to adapt to context without the model inventing layout, styling or behaviour. It is in early access and not yet generally available.

How does Weft keep AI-composed screens within the rules?

By design, the model can only choose from a catalogue of components with typed properties. Each composed screen is checked against a schema and semantic rules before it is shown, every screen served is recorded so it can be reproduced, and a surface can be pinned back to a hand-authored fallback. The checks apply to a screen's structure. They do not judge whether it is the best screen for a user, and they do not replace testing the components themselves.

Where is Reply Intelligence Group based?

Reply Intelligence Group is based in Watford, Hertfordshire, and works with businesses across the UK. Reply Intelligence Group Ltd is registered in England and Wales; its registered office is 71-75 Shelton Street, Covent Garden, London WC2H 9JQ. Products are built UK-first around GDPR posture, UK appointment economics and operator workflows.

Looking for implementation work rather than a product?