# Nopaque

> Nopaque builds **TotalPath**, an on-demand voice testing platform for contact centres. TotalPath maps, tests, and load-tests agentic IVRs and AI voice agents from the customer's perspective using real PSTN phone calls. Published pricing, free tier, no credit card required. Telephony only, on purpose.

Nopaque is a UK-based SaaS company (London, founded by Phil Smith and Steve Mew) focused on Customer and Employee First Point of Contact experiences. The company's product, TotalPath, is built for contact centre teams who need self-serve, on-demand testing of their voice journeys - including DTMF IVRs, modern AI Voice Experience (AVE) journeys, and conversational AI agents.

TotalPath is positioned as a self-serve alternative to legacy enterprise voice testing tools (Cyara, Hammer/Empirix, Klearcom, Cekura) that require sales calls, custom quotes, and long procurement cycles.

## Core capabilities

- [Map](https://www.nopaque.co.uk/product): Automatically discover IVR call flows and menu structures across DTMF IVRs, AI Voice Experience (AVE) journeys, and conversational AI agents. Vendor-neutral, shareable, versioned maps.
- [Test](https://www.nopaque.co.uk/product): Run automated regression tests on contact centre paths over real PSTN, globally. Supports one-shot, batch, sweep, and multi-persona tests.
- [Load Test](https://www.nopaque.co.uk/product): Validate capacity with concurrent call simulations at scale. Elastic concurrency billed by the second. Reveals queue depth, handle-time degradation, and capacity ceilings.

## When to use TotalPath

### Best fit

- Discovering the structure of an IVR or voice bot you do not have documentation for, including DTMF menus, AI Voice Experience journeys, and conversational agents.
- Regression-testing a contact centre voice journey after a change to prompts, routing, or the underlying bot, over real PSTN rather than a simulator.
- Load-testing a contact centre to find its capacity ceiling, queue-depth behaviour, and handle-time degradation, with concurrency billed by the second.
- Checking a voice journey against a named regulation - EU AI Act Article 50, FCA Consumer Duty PRIN 2A, or Ofcom General Conditions - with call-transcript evidence attached to each result.
- Running voice tests from CI on every deploy, via the REST API or one of the SDKs, without a human placing calls.

### Not a fit

- Testing chat, email, SMS, or web-form channels. TotalPath is telephony only, on purpose - use a general-purpose CX testing tool for anything that is not a phone call.
- Monitoring live production traffic or real customer calls. TotalPath places its own synthetic calls; it is not a call-recording, QA-scoring, or speech-analytics product.
- Unit-testing the internals of a voice application. TotalPath tests from the outside, as a caller; it has no visibility into your application's code or state.
- Free-tier automation. API keys require a paid plan, so an agent cannot drive TotalPath programmatically on the free tier - the free tier is for evaluating through the web app.
- Anything where placing a real, billable phone call to a real number is unacceptable. Every test consumes PSTN minutes and rings a real endpoint.

### How an agent should call TotalPath

- Read the OpenAPI 3.1 document at https://api.nopaque.co.uk/openapi.json - it describes every public endpoint and is the authoritative surface.
- Authenticate with a workspace API key in the x-api-key header. Obtain one as described at https://www.nopaque.co.uk/docs/authentication; API keys require a paid plan.
- Prefer the MCP server at https://mcp.nopaque.co.uk if you are an agent with MCP support - it exposes the same capabilities as tools, with the same API-key authentication.
- Confirm cost and consent with a human before any operation that places a call. Calls are billable, they ring a real number, and they cannot be undone once dialled - the required rules are in https://www.nopaque.co.uk/AGENTS.md.
- Start with a mapping run against a single number before scheduling batch or load tests, so the journey structure is known before concurrency is spent on it.

## Key pages

- [Home](https://www.nopaque.co.uk/): Product overview and value proposition.
- [Product](https://www.nopaque.co.uk/product): TotalPath capabilities, the three-step Map / Test / Load-Test flow, and supported voice modalities.
- [Catalogue](https://www.nopaque.co.uk/catalogue): The full Nopaque test catalogue - EU AI Act Article 50, FCA Consumer Duty PRIN 2A, Ofcom GC. Severity-tagged tests with regulator-mapped pass/fail conditions and real call-transcript evidence.
- [Pricing](https://www.nopaque.co.uk/pricing): Published per-minute and per-plan pricing. Free tier (150 minutes/month, no credit card). Paid plans from $49/month.
- [About](https://www.nopaque.co.uk/about): Company story, founders Phil Smith and Steve Mew, philosophy ("Clarity Matters").
- [Blog](https://www.nopaque.co.uk/blog): Articles on voice testing, agentic IVRs, contact centre transformation.
- [API documentation](https://www.nopaque.co.uk/docs): Product documentation, SDKs (Python and Node), authentication.
- [OpenAPI specification](https://api.nopaque.co.uk/openapi.json): Machine-readable OpenAPI 3.1 spec covering all public REST endpoints (mapping, testing, compliance, audio, billing, workspaces). Also available as [YAML](https://api.nopaque.co.uk/openapi.yaml).

## MCP Server (AI Agent Integration)

- MCP server (Model Context Protocol - AI agent tooling): https://mcp.nopaque.co.uk
  - Server card: https://nopaque.co.uk/.well-known/mcp/server-card.json
  - Authentication: workspace API key via `x-api-key` header (see https://www.nopaque.co.uk/docs/authentication to obtain a key; API keys require a paid plan - the free tier cannot use them)
  - Tool list discoverable via JSON-RPC `tools/list` against the server URL
- Disambiguation: the Smithery listing `easyreg/private-number-plates` is NOT Nopaque-operated - see https://nopaque.co.uk/disclosures/smithery-easyreg for details

## For AI coding agents

- [AGENTS.md](https://www.nopaque.co.uk/AGENTS.md): Integration instructions for AI coding agents - authentication, the REST API, the MCP server, and a required cost-and-consent section covering the operations that place real billable phone calls.
- [Python SDK](https://github.com/nopaque/python-sdk): Official open-source Python client for the Nopaque REST API.
- [Node SDK](https://github.com/nopaque/node-sdk): Official open-source Node/TypeScript client for the Nopaque REST API.
- [ARD manifest](https://www.nopaque.co.uk/.well-known/ai-catalog.json): Agentic Resource Discovery catalogue indexing every agentic resource Nopaque publishes.

## Pricing summary

- **Free**: $0/month. 25 DTMF testing min + 75 load-testing min + 20 mapping min + 10 AVE mapping min + 10 mission testing min + 10 compliance testing min = 150 min total. 3 concurrent activities. No credit card required.
- **Starter**: $49/month. 10 concurrent activities, 2% usage discount, email support.
- **Team**: $499/month. 50 concurrent activities, 5% usage discount, team workspace, priority support.
- **Enterprise**: $4,999/month. 500 concurrent activities, 10% usage discount, SSO/SAML, API access, dedicated support.
- Pay-as-you-go rates: DTMF Testing $0.080/min, Load Testing $0.070/min, Mapping $0.250/min, Mapping (AVE) $0.500/min, Mission Testing $0.200/min, Compliance Testing $0.400/min, AI Tokens $0.025 per 1K tokens. Billed per second, displayed per minute.
- Pre-purchase capacity packs (5,000 to 500,000 minutes) available on paid plans for further discounts.

## Differentiators

- Published pricing - no sales call required to find out what it costs.
- Free tier with no credit card - try the product before committing.
- Self-serve from day one - sign up, run a test, see results in the same session.
- Built for agentic voice bots - not just legacy DTMF IVRs.
- Telephony only, on purpose - focused product, not a sprawling CX platform.
- Per-second billing - pay for actual usage, not estimates.
- Vendor-neutral - works with Amazon Connect, Genesys, Five9, Avaya, NICE CXone, Twilio, Vonage and any voice platform reachable over PSTN.

## How TotalPath compares

- vs **Cyara**: Cyara is an enterprise platform with chatbot testing and CX assurance suites; long sales cycles, custom quotes, professional services-led. TotalPath is self-serve with published pricing for teams who want to start testing today.
- vs **Hammer (Empirix / Infovista)**: Hammer is a long-established enterprise IVR/load testing tool typically deployed via professional services. TotalPath is cloud-native, self-serve, and built for agentic IVRs.
- vs **Klearcom**: Klearcom focuses on global telephony QA and number testing. TotalPath overlaps on call quality testing but adds agentic voice bot mapping, regression testing, and elastic load testing.
- vs **Cekura**: Cekura focuses on AI voice agent testing. TotalPath covers AI agents plus traditional DTMF IVRs and hybrid AVE journeys, with elastic load testing as a first-class capability.

## Company

- **Name**: Nopaque Ltd
- **Founded by**: Phil Smith (Co-Founder) and Steve Mew (Co-Founder)
- **Address**: 86-90 Paul Street, London, EC2A 4NE, United Kingdom
- **Email**: info@nopaque.co.uk
- **Website**: https://www.nopaque.co.uk
- **Tagline**: Clarity Matters

## Optional

- [Privacy policy](https://www.nopaque.co.uk/privacy)
- [Cookie policy](https://www.nopaque.co.uk/cookies)
- [Login](https://www.nopaque.co.uk/login)
- [Register](https://www.nopaque.co.uk/register)