The platform · one page

Multi-agent. Multi-channel.
Multi-tenant.

The unified AI agent platform — without the enterprise contract.

Multi-agent orchestration, multi-harness runtime variants, run-state continuity, contextual RAG, voice + telephony, and background automation — behind a self-service admin dashboard, on open standards. Self-hostable end-to-end. Built in Vienna.

0+MCP connectors
0Retrieval stages
0%Self-hostable
The defensible claims

Sixteen things most platforms can't say.

Every row is a claim you can check — with the evidence beside it. No marketing fog.

ClaimEvidence
01Unified text + voice + telephonyWidget, voice widget, Twilio / Vonage / FreeSWITCH — all first-class in-product, same agent graph.
0210,000+ integrations, one clickMCP Catalog aggregating Docker / Official / Smithery registries. Salesforce, HubSpot, Zendesk, Jira, Slack, SharePoint already there.
03Best-in-class document processingSelf-developed doc2md library with vision-model image augmentation + 4 scraping engines with Playwright automation.
04Premium chat widget, not a bubbleCarousels, detail cards, action pills, follow-ups, voting, markdown + code, image upload/lightbox, voice mode — fully themeable.
05Full headless APIComplete REST + SSE chat API + full KB CRUD/push/query API. Drive ChatFlow as a pure reasoning backend if you want to.
06Self-host your whole stackBGE-M3 + Qwen3-Reranker + Qwen3-ASR/TTS + Silero VAD + local LLM. Zero external data.
07Agency-grade multi-tenancy6-level per-tenant RBAC, god-mode, snapshots, in-product billing with EU VAT and Stripe. One install, ten clients.
08Contextual retrieval built-inAnthropic-style LLM-generated chunk prefixes + hybrid dense/BM25 + reranker. Rare outside enterprise-tier.
09True multi-agent with parallel consultsHandoffs (control transfer) and silent consult edges executing in parallel. Not a single fine-tuned "bot".
10Multiple harnesses per tenantA tenant can run separate bot applications for support, sales, staging, voice, or campaign variants — each with its own graph scope, triggers, memory, analytics, and runtime instructions.
11Run State keeps long work coherentSchema-validated workflow state with per-stage read/write policies. Persists across handoffs, pipeline stages, fanout branches, waits, approvals, and resumptions — survives server restarts and graph edits via pinned compiled-graph snapshots.
12Deterministic workflow orchestrationPipeline · condition · error edges; transform · state-update · router · join · wait · poll-until · approval · output nodes. Graph-driven control flow with durable pause/resume and an Operations queue for human approvals — not just model-chosen handoffs.
13Build harnesses by chattingHarness Architect — a chat-based wizard backed by a privileged System MCP server — interviews you, recommends agentic / pipeline / hybrid, generates a v3 import package, validates and previews, and applies only after explicit confirmation plus an automatic backup snapshot. Replaces the entire v1 wizard zoo.
14Instructions from external sourceBind agent instructions to a file, repo path, or external doc. Versioned, auditable, sharable across harnesses — no copy-paste prompts drifting between environments.
15Open standards, no lock-inOpenAI Agents SDK, MCP, BGE-M3, Qdrant, OAI-compatible endpoints, snapshot JSON export.
16Proven at scale since 2023AI-Concierge predecessor — 20k+ monthly users across tourism, municipal, and enterprise deployments.
Versus the market

Where ChatFlow sits.

Nineteen capability rows across six buckets of competitor. ChatFlow column highlighted.

Chatbase / TidioVoiceflow / BotpressIntercom Fin / AdaKore / CognigyVapi / RetellChatFlow
Multi-agent graph+ parallel consults
Multi-harness runtime variantsprod / staging / channel variants
Persisted run stateinspect + resume
Deterministic graph orchestrationpipeline · condition · approval
Source-bound agent instructionsgit / file / external doc
Contextual retrieval (LLM prefixes)
Hybrid + reranker3 backend modes
Voice + telephony unifiedself-host + FreeSWITCH
MCP-native10k+ catalog
Self-hostable stack
True multi-tenant (agency-ready)
Billing in-product (VAT, Stripe)n/a
Structured KBs + carousels
GDPR + EU AI Act mapping
Regression tests w/ LLM Judge15 languages
Triggers / scheduled automation
Audit-grade release managementversioning + AI changelog + customer notes
User MFA + tenant-enforceable policyTOTP + recovery codes
Price tier$40–500/mo$50–800/mo$0.99/res+$100k+ ACV$0.05/min~5–15% of Cognigy
shippedpartialmissing
The platform, layered

Nineteen layers, one product.

Each layer is a full product in its own right. Tap any row to unpack it.

Channels
Widget · voice · phone · triggers · REST
Web widget (<chatbot-widget>, ~54KB gz), voice widget, SIP telephony (Twilio / Vonage / FreeSWITCH), REST + SSE chat API, Triggers (cron + inbound webhooks with vendor presets for Zendesk, Intercom, HubSpot, Jira, Slack, Freshdesk).
Identity / Persona
9-layer composed prompt, Persona Wizard
Persona + Memory + Date + Agent + Skills + KB Guidance + Citation + Image + Style. Website-scanning Persona Wizard, multilingual greetings with memory-aware personalization.
Agent Graph
Handoffs + consults + deterministic orchestration
Directed graph with two model-chosen edge types (handoff, consult) and three deterministic orchestration edges (pipeline, condition, error). Orchestration nodes for transform, router, join, state-update, wait, poll-until, approval, and output. Per-agent model/temperature/reasoning, Structured-Output mode (schema builder → Pydantic), and a single chat-driven Harness Architect — a system-MCP-backed wizard that interviews, designs, validates, previews, and applies entire harnesses with backup snapshots. Replaces the old per-agent / per-skill / Assess-Graph modals from v1.
Harnesses
Multiple runtime variants per tenant
A harness is a runnable application boundary inside a tenant. A single tenant can run several side-by-side: support, sales, staging, voice, campaign — each with its own selected agent graph, graph-wide instructions, scoped triggers, scoped memory and analytics, and cloneable variants. Tenant assets stay shared; runtime behavior becomes selectable, swappable, and auditable per harness.
Run State
Durable, schema-validated workflow state
Run-scoped state with a tenant-defined schema, read/write policies per agent and per edge, deterministic state-update nodes, projection across handoff/consult/pipeline edges, trigger-payload seeding, and tool-result-to-state mapping. Long workflows stay coherent across handoffs, resumptions, waits, and approvals — without leaning on chat history.
Instruction Sources
Bind agent prompts from file, repo, external doc
Agent instructions can be sourced from external systems instead of typed inline. Version-pinnable, auditable, and sharable across harnesses. Edit prompts in your own SCM, deploy with the same review/PR process as code. The admin still shows the resolved final instructions and the source of record.
Skills
Reusable workflow bundles
Instructions + tool refs + KB refs attached to agents. Platform-level skill templates. AI-generated topology-aware instructions.
Tools
Builtin · webhook · MCP · 10k+ catalog
@register_builtin code tools, webhook tools with /schema auto-discovery + {{secret}}/{{var}}/{{param}} templating + retries, MCP servers (SSE / Streamable HTTP / stdio) with auto-discovery, and the MCP Catalog with 10,000+ one-click installable servers.
Knowledge
4-stage retrieval + structured KBs + doc2md
Prose KBs: BGE-M3 dense + Qdrant BM25 + RRF + Cohere/Qwen3-Reranker + LLM-generated context prefixes. Structured KBs: SQLite ingest with auto-detected column roles, filtered+semantic hybrid query, side-channel carousel/detail-card rendering, Push API with SHA-256 change detection. doc2md with AI image augmentation. Multi-engine scraper (Trafilatura / Crawl4AI / Hybrid / LLM Extraction).
Chat Widget
Carousels, pills, voice, sign-in — fully themeable
Standalone Web Component (~54KB gz). Rich artifact vocabulary: carousels, detail cards, action pills, follow-ups, inline feedback, GFM markdown with syntax highlighting, image upload/paste/drag-drop with lightbox, voice-mode UI, and built-in sign-in (SSO passthrough via auth-callback attribute or email-OTP magic-link with localized panel in 12 languages). Allowed-origins enforcement.
Developer API
Full REST + SSE headless engine
Complete REST + SSE chat API. Full API control over knowledge bases (CRUD, push with SHA-256 change detection, rebuild, query with scoring).
Memory
Session · Facts · Episodes
Redis session (~1 ms, 2 h TTL) + Facts (Mongo+PG, confidence + importance decay + supersede history) + Episodes (Mongo+Qdrant semantic). Per-agent scope + category filters. GDPR purge/export.
Guardrails
LLM · pattern · webhook — per direction
LLM classifier, pattern match, webhook validator. Per-agent, per-direction (input/output), kill-switch.
Voice
OpenAI Realtime or self-hosted Qwen3
Dual backend: OpenAI Realtime (WebRTC, 13 voices, gpt-realtime-1.5) or self-hosted (Silero VAD → Qwen3-ASR → full Agents SDK with multi-agent handoffs → Qwen3-TTS).
Telephony
FreeSWITCH · Twilio · Vonage
FreeSWITCH + mod_audio_fork (self-host, with PBX extension transfer), Twilio, Vonage. Automatic end_call + transfer_call injection, schedule-aware transfer targets, inactivity monitor, forced server-VAD for codec-compressed audio.
Model Registry
Per-tenant, role-based, reasoning-aware
Per-tenant, role-based (Agent Default / Fast / Smart / Nano / Context-Prefix / Scraping / Memory-Extraction / Voice / Wizards). Presets for OpenAI, Azure, Anthropic, Google, Mistral, Qwen. Reasoning-param translation across 5 providers. vLLM / Ollama ready.
Platform
Multi-tenant · RBAC · SSO · MFA · snapshots · audit · release mgmt
Multi-tenant isolation at DB + Qdrant + Redis + cache + API. 6-level RBAC per-tenant. Three-layer auth: API keys (scoped) for programmatic access, JWT for admin dashboard, end-user OIDC passthrough (Entra / Okta / Keycloak / Auth0 / Google) or email-OTP magic-link for widget users — with group-based KB ACLs on top. Per-user TOTP MFA with single-use recovery codes and tenant-enforceable policy (off / admins / all). Explicit operator modes (Platform vs Tenant) with audit-tagged impersonation. Configuration snapshots (version, restore, cross-platform portability). Built-in release management: git-describe versioning, deploy history, AI-generated weekly project history, draft → review → publish releases with GitHub auto-tag. Audit log with CSV export. Fernet-encrypted secrets.
Billing
Tiered · flat · token · versioned with grandfathering
Tiered / Flat / Token-passthrough / Free plans, versioned with grandfathering, EU-VAT + reverse-charge handling, Stripe + manual workflows, 3-level dunning with auto-reactivation, DATEV/BMD CSV export, Jinja2 customizable templates.
Observability
Inspector · Playground · Run State · Analytics · regression tests
Conversations inspector (Messages / Events / Retrieval w/ chunk scores / Memory / Trace), Playground with SSE stream, persisted Trace, Run State and Memory inspectors, Analytics with anomaly banners, Prometheus + Grafana + 7 pre-configured alert rules, regression-test framework with LLM Judge in 15 languages.
Compliance
GDPR · EU AI Act · self-host
GDPR role matrix, Subject-Rights endpoints, retention automation (30/90/180/365 d), ROPA template, EU AI Act per-Article mapping, Shared Responsibility Model doc, fully self-hostable pipeline.
Beyond handoff

Deterministic graph orchestration.

Handoff and consult are model-chosen. Pipeline, condition, and error are runtime-chosen. Both live in the same graph — handoff for collaboration, orchestration for control flow.

Example · refund workflow
handoff
structured_json
intent = refund
else
pipeline
Triage
Extractor
Router
Validator
Approval
Support

Edges

Two model-chosen, three runtime-chosen.

  • handoff
    model-chosen

    Source agent transfers control. Optional run-state projection.

  • consult
    model-chosen

    Source agent calls a target agent as a tool. Parallel-capable.

  • pipeline
    runtime-chosen

    Target runs automatically when source completes. Auto, structured-JSON, or text payload.

  • condition
    runtime-chosen

    Branch on previous output, run state, trigger payload, or metadata.

  • error
    runtime-chosen

    Route on validation, tool, model, or timeout failures.

Nodes

Agent nodes plus seven deterministic node types.

  • agent

    LLM agent. Per-agent model, tools, structured-output schema.

  • transform

    Deterministic JSON-to-JSON or JSON-to-text mapping. No LLM.

  • router

    Centralized branching point. Easier to read than fan-out from a single agent.

  • state_update

    Schema-validated write into run state. No LLM.

  • output

    Explicit terminal delivery — conversation, artifact, webhook, or email.

  • join

    Wait for parallel branches and merge results.

  • wait

    Pause durably for time, datetime, or external event.

  • poll_until

    Active deterministic readiness check (e.g. uploaded-document status). Loops with caps until success, failure, or timeout — never lets an agent guess infrastructure state.

  • approval

    Human-in-the-loop gate: approve, reject, or edit payload. Operations queue surfaces the task with claim/lock semantics.

How a request flows

From keystroke to reply, step by step.

Every request passes through the same eleven stages — whether it comes from a widget, a phone call, or a webhook.

Step 01
Arrive
Widget / API / phone → resolved to tenant via API key, phone number, or origin.
Step 02
Select harness
Default, API-selected, widget-selected, or trigger-selected harness fixes the runnable graph, graph-wide instructions, memory scope, and analytics scope.
Step 03
Resolve instructions
For agents bound to an external source (file, repo, doc), instructions are fetched at the pinned version. The composed prompt is the same shape, but the source of record stays out of the admin.
Step 04
Input guard
Pattern → LLM classifier → webhook validator. Per-agent, per-direction.
Step 05
Load context
Conversation history, rolling summary, harness-scoped memory context (facts + episodes, agent-scoped), and current run-state projection per stage.
Step 06
Reason
Entrypoint agent reasons. Calls tools. Optionally consults specialists in parallel. Optionally hands off. Deterministic orchestration edges (pipeline, condition, parallel/join, wait, approval) drive control flow when configured.
Step 07
Retrieve
4-stage retrieval: dense + BM25 + RRF + rerank. Context prefixes injected.
Step 08
Update run state
Structured run state records step progress, decisions, resumptions, and inspectable debug checkpoints. Deterministic state-update nodes write schema-validated values; agent-driven writes use read/write policies.
Step 09
Output guard
Response validated. Then SSE stream to the client — with side-channel events for carousels and detail cards.
Step 10
Persist
Async memory extraction, rolling-summary update, conversation and run state persisted, metrics incremented.
Step 11
Phone extras
Codec conversion, barge-in, inactivity monitor, transfer-schedule evaluation.
4-stage retrieval

Contextual, not just semantic.

Every chunk gets an LLM-generated context prefix before embedding. Anthropic-style. Rare outside enterprise-tier.

Stage 1
Encode
Query → BGE-M3 dense (1024d). Self-hosted, micro-batched.
1024d
Stage 2
Hybrid search
Dense + Qdrant server-side BM25 · RRF fusion · top 100.
top_k = 100
Stage 3
Rerank
Cohere rerank-v4.0-pro or Qwen3-Reranker · top 15.
top_k = 15
Stage 4
Format
Chunks with context prefixes → injected into agent context.
context +prefix
Use cases

What customers actually build.

Seven real patterns, each with the problem, the current stopgap, the ChatFlow build, and the value delivered.

Medical / Dental practice

Problem
Front desk overwhelmed. Emergency calls mixed with rescheduling. Multilingual patients. After-hours gaps.
Today
IVR trees (rigid) + human reception (bottleneck) + appointment webforms (drop-off).
ChatFlow
Triage agent → Appointment (eTermin webhook) + Insurance (structured fee schedule) + Medical FAQ + Emergency Router (FreeSWITCH transfer). Voice on phone. Multilingual.
Value
~80% inquiry deflection. Emergencies reach human in seconds. 24/7 booking. Staff reclaimed for complex work.

E-commerce advisor

Problem
5k+ SKUs. Repetitive 'where is my order?'. Product-selection friction.
Today
Chatbase-type bot (can't act), Zendesk macros (human-gated), separate order-status page.
ChatFlow
Shopping Assistant → Product Advisor (structured KB catalog + carousels) + Order Support (tracking / returns webhooks) + Technical Support (docs KB) + MCP to WMS.
Value
Actions happen inside chat. Industry typical: 40–60% ticket deflection, CSAT lift from carousel discovery.

B2B SaaS support

Problem
Complex product. Developers want API specifics, admins want config, end-users want how-to.
Today
Separate docs + ticketing + Discord. Context-switching cost.
ChatFlow
Support Triage → Developer (API KB + MCP internal docs + Jira) + Admin (config KB) + End User (how-to FAQ) + Status (status API). /schema endpoints auto-discovered.
Value
Every new microservice becomes an agent capability overnight. Ticket volume down, dev activation up.

Real estate advisor

Problem
Multilingual clients. Property data in CRM. Viewing scheduling. Financing questions across jurisdictions.
Today
Website forms + human agents by phone + financing calculators.
ChatFlow
Structured KB = property catalog (20+ filterable columns + semantic). Viewing webhook. Mortgage KB per country. Voice with EU residency via Vonage.
Value
Clients search by voice in native language, see properties as carousel, book viewing — without leaving the conversation.

Internal knowledge hub

Problem
Info scattered across Confluence/SharePoint/HR portals. Employees lose hours.
Today
Enterprise search (Glean) at $40–50/seat + Copilot (generic).
ChatFlow
Specialist agents per domain (HR / IT / Finance / Product), each with their own KB. IT agent has MCP to AD + ServiceNow. Per-role guardrails redact salary/PII.
Value
Domain-scoped, action-capable (create tickets, check AD membership), isolated per department via multi-tenant.
Proven · 20k+ MAU

Tourism concierge

Problem
Guest needs span activities, dining, room service, logistics across multi-country properties.
Today
Concierge desks + per-property apps.
ChatFlow
Per-property tenant, structured KBs for activities/restaurants, room-service webhook, voice widget in-room app. Parent brand shares skill templates.
Value
Proven pattern. GoodGuys' AI-Concierge predecessor serves 20k+ monthly users across Basel, Zurich, Salzburg, Vienna, Villach and the Dolomites.

Back-office automation

Problem
Email triage, daily digests, lead qualification from Typeform, document comparison — work that doesn't need a chat UI.
Today
Zapier / Make + custom code + separate LLM calls.
ChatFlow
Trigger fires (cron or webhook) → same agent graph runs → result returned sync, via callback, or written into conversation. Vendor presets for Zendesk/Intercom/HubSpot/Jira/Slack.
Value
One configuration serves both live conversations and unattended automation. No duplicate infra.
Ecosystem

Sits on top of the stack you already run.

Twelve categories of integration. Tap a method chip to highlight all integrations using it.

Ticketing / Helpdesk
ZendeskFreshdeskIntercomJira
webhook
CRM
HubSpotSalesforcePipedrive
MCP
Scheduling / Calendar
eTerminCalendlyGoogle CalendarMS Calendar
webhook
Knowledge / Docs
ConfluenceSharePointNotionPublic websites
API
Identity / SSO
Microsoft Entra IDOktaKeycloakAuth0Google WorkspaceAWS Cognitoany OIDC providerEmail OTP
native
Voice / Telephony
TwilioVonageFreeSWITCH PBX
native
Payments
Stripe
native
Analytics / Monitoring
PrometheusGrafanaPagerDutySlack
native
Data pipelines
CI/CDETLRegression tests
API
Messaging
WhatsAppSlackTeamsEmail
webhook
Compliance
DATEVBMDVAT validators
native
AI / LLM providers
OpenAIAzureAnthropicGoogleMistralCoherevLLMOllama
native
Enterprise-ready sign-in

Know who's on the other side of the chat.

Two independent flavors, enable either or both from the admin dashboard. The widget forwards the verified identity on every request — conversations get a real end-user record, KB access can follow group claims, and GDPR delete lands on a real row.

Flavor 1

OIDC Passthrough

Your host page is already logged in. It hands a fresh JWT to the widget via a small callback; the backend verifies against your IdP's JWKS. No client registration on our side, no redirects, no refresh tokens leaving your perimeter.

  • Works with any OIDC-compliant provider — JWKS-URL auto-discovered from the issuer
  • Algorithm allowlist (RS/ES/EdDSA). HS* and alg:none always rejected
  • Claim mapping supports dot paths and namespaced claims
  • Three modes: disabled · optional · enforced
Flavor 2

Email OTP (Magic-link)

No IdP required. User types their email, receives a 6-digit code, stays signed in via a secure cookie. Same entry point whether they're new or returning — no separate signup flow.

  • Redis-backed codes, rate-limited per email and IP
  • HttpOnly+Secure cookie, tenant-bound, 30-day sliding
  • Optional hCaptcha and tenant-customisable consent copy
  • DELETE /end-user/me endpoint for GDPR Art. 17
Verified against
Microsoft Entra IDOktaKeycloakAuth0Google WorkspaceAWS Cognitoany OIDC-compliant IdP

Group-based KB ACLs

Once users sign in with an IdP that emits group claims, restrict individual Knowledge Bases to specific groups. Non-members don't get a "permission denied" — the KB is silently dropped from retrieval, so they can't tell it exists.

Visible where you need it

End-user email and auth-method show up in the admin conversation list, with full identity + groups in the meta tab of each conversation.

Agentic access to company systemsOn the roadmap · In active development

Let the agent act on your company systems — safely.

A centralized Connections layer gives every ChatFlow feature the same OAuth, refresh, audit, and revocation plumbing — so new integrations never come at the cost of new security surface. Enterprise mail is the first consumer; Drive, Calendar, CRM and more follow on the same rails.

Foundation

One OAuth layer, every integration

Instead of bolting auth onto each feature, one infrastructure layer owns authorize / callback / refresh / revocation / audit. Every future integration inherits the same admin UX, security controls, and operator visibility.

  • Capability-scoped tokens — agents receive narrow, auditable access, never raw credentials or refresh tokens
  • Incremental consent — one connection per account powers many features; adding a feature asks only for its additional scopes instead of spinning up a new sign-in
  • Revocation cascade — one revoke pauses every dependent feature; admins see one unified banner, not per-feature noise
  • Encrypted at rest (KMS-ready envelope design), per-fetch audit log, PKCE + signed state on every OAuth flow
First consumer

Enterprise mail: inbound, outbound, threaded

Microsoft 365 and Google Workspace via XOAUTH2 through the Connections layer, plus self-hosted SMTP/IMAP and Resend for transactional sends. Inbound mail triggers agent runs through the existing trigger pipeline; replies thread naturally; attachments can be promoted into KBs under strict ACL control.

  • XOAUTH2 for Entra ID / Google Workspace — no passwords ever stored by us
  • Threaded conversations via plus-addressing + In-Reply-To / References fallback
  • Three-gate attachment-to-KB ingestion (agent scope → KB opt-in → optional per-doc ACL narrowing)
  • DMARC-fail drop, bounce classification, idempotent retries, dry-run mode for test environments
Designed for
Microsoft 365 / Entra IDGoogle WorkspaceSelf-hosted IMAP / SMTPResendMicrosoft Graph (next)Salesforce · HubSpot (next)Slack · Teams (next)any OAuth 2.0 service

Forensics-ready by design

Every token fetch joins back to the originating agent run. Every email-ingested KB document carries provenance to the original message. A single GDPR deletion walks the chain — chunks in Qdrant, text in MongoDB, blobs in object storage — from one inbound-email ID.

Scope-gate your agents

Agents never see raw tokens, only tool results. Capability gating is enforced at the token broker; per-agent RBAC decides which mailboxes can be sent from and which KBs can be written to. Platform operators can pause any provider centrally via a feature flag — no per-feature hunting, no code change.

Early-access pilot partners welcome during v1 — get in touch if you'd like your enterprise mail or data systems on the integration roadmap.

Strengths & gaps

Everything we ship, and what we don't.

Most competitor sites won't put this on one page. We will.

Strengths

  • 4-stage contextual retrieval with 3 reranker backends — enterprise-grade in one product.
  • Full vertical coverage: multi-agent + RAG + voice + telephony + triggers + billing + compliance.
  • 10,000+ integrations via the MCP Catalog — one-click Salesforce, HubSpot, Zendesk, Jira, SharePoint.
  • Self-developed doc2md with AI image augmentation + 4 scraping engines with Playwright.
  • Production-grade chat widget — carousels, pills, feedback, markdown, voice. A product in itself.
  • Full headless API — drive ChatFlow as a pure reasoning backend if you want.
  • Self-hostable end-to-end (BGE-M3 + Qwen3 + Silero + local LLM). Zero external data mode.
  • True multi-tenancy with god-mode, snapshots, in-product billing with EU VAT.
  • AI-Concierge in production since 2023. 20k+ monthly users across tourism, municipal, enterprise.
  • Open standards: OpenAI Agents SDK, MCP, Qdrant, OAI-compatible endpoints. Snapshot JSON export.
  • Observability as a feature: chunk-level scores, memory inspector, regression tests with LLM Judge.
  • Enterprise-ready sign-in: OIDC passthrough for Entra / Okta / Keycloak / Auth0 / Google, plus email OTP for public widgets. Group-based KB ACLs and a GDPR delete endpoint — all self-service in the admin.

Known gaps

  • Brand recognition — no Fortune-500 logo wall yet. Credibility gap for US enterprise deals.
  • No SOC2 Type II / ISO 27001 yet. Table stakes for US enterprise procurement.
  • No outcome-based pricing shipped yet. Market trending per-resolution for CX.
  • Native channel adapters (WhatsApp, Messenger, Slack) still need webhook wiring. MCP servers exist, one-click UX does not.
  • Limited analyst coverage. No Gartner MQ placement, no public Tau-bench score.
  • LangSmith-depth eval features missing: call replay, A/B traffic splitting, dataset eval matrices.
  • Curated agent/skill template marketplace is new vs Voiceflow/Botpress libraries.
  • SAML direct bridging is offloaded — the host page must exchange SAML for a JWT before handing it to the widget.
  • Art. 18 GDPR restriction only supported as workaround.
  • Consent management offloaded to tenant integration layer.
Who it's for

Five audiences, ranked by fit.

If you're in one of these buckets, ChatFlow is built for you. If you're not — we'll tell you.

01

EU mid-market in regulated industries

Why you
Healthcare, public sector, real estate, hospitality. Need GDPR + AI Act conformance. Can't afford Cognigy. Outgrowing Chatbase.
Not for you if
US-first companies with no EU residency needs.
02

Digital agencies and white-label resellers

Why you
True multi-tenancy, per-tenant billing, snapshots, reseller economics. Ten clients on one install.
Not for you if
Agencies that want to resell someone else's frontend.
03

Tourism, hospitality, municipal assistants

Why you
Direct extension of AI-Concierge market. Proven via existing deployments. Multilingual from day one.
Not for you if
Single-property boutique operators with no IT.
04

B2B SaaS with deep product integrations

Why you
Unified developer / admin / end-user support via MCP. Every microservice becomes an agent capability.
Not for you if
Teams that want a pre-trained support bot with no configuration.
05

Internal-ops teams

Why you
Glean-style knowledge assistants with action-taking and domain isolation — without $50/seat for 500 employees.
Not for you if
Organizations where IT wants one vendor across all productivity (Copilot territory).
Messaging

Six things to remember.

Beyond RAG. Beyond chatbots.
Every channel. One platform.
Own your stack.
Deterministic orchestration, not magic routing.
Enterprise governance at ~5–15% of Cognigy.
Built for agencies and regulated mid-market.

Try it. Three ways.

Live now · ready

Talk to a live bot

Click to talk

Talk to ChatFlow about ChatFlow. The voice stack, live.

Try asking:«How does contextual retrieval work?»

Live demo. Audio is processed in real time. 3-minute cap. privacy.

Mic muted while Flo speaks — prevents feedback.
Transparent · EU VAT

See pricing

Review how ChatFlow is priced per tenant, how packaging scales with usage, and how EU VAT and reverse-charge are handled out of the box.

  • Per-tenant, usage-based pricing
  • EU VAT + reverse-charge built-in
  • No lock-in — export your data
1:1 · 20 min with the team

Book 20 min

Mon4
Tue5
Wed6
Thu7
Fri8
10:0014:30

Walk through your exact use case with us, live in the product. No slides, no sales deck — just your stack against ours, and honest answers.

  • Your concrete use case
  • Live inside the product
  • Technical Q&A, no gatekeeping