Give your producta mind.
AI that answers questions across your website, Slack, phone calls, and developer tools — grounded in your own documentation.
Free for 7 days · Cancel anytime
Trusted by innovative teams like
Train once. Deploy everywhere.
Your knowledge base powers every surface — chat, inline, Slack, voice, and IDE. Build it once and your AI shows up wherever your users are.
The founder's knowledge, always in the room.
A calm, competent intelligence that lives inside your product. Users never feel lost, and they never have to leave your app to understand it.
- Answers grounded in YOUR knowledge, not generic AI
- Every response includes citations you can verify
- Understands messy, vague questions and extracts the real intent
- Asks one smart clarifying question instead of guessing
- Floating and embedded modes to fit any layout
Highlight to understand.
Users highlight a setting, an error, or a concept, and get an instant, cited explanation from your knowledge base. No searching. Just understanding.
Voice support that never sleeps.
Customers call a real phone number. Your AI answers using the same knowledge base — no scripts, no hold music.
- Phone numbers in 60+ countries — local and toll-free
- Same knowledge base as chat — train once, works everywhere
- Human escalation — transfers to your team when needed
- Call history, transcripts, and recordings in your dashboard
- Available as a Growth or Scale add-on
Tag @pernoex in any Slack thread.
It reads the full conversation — including screenshots — and answers grounded in your docs. No context switching, no ticket filing.
- Reads full thread context including images
- Same knowledge base — no separate training needed
- Custom Slack bot on Scale — your branding, your name
- Works in public channels, private channels, and DMs
Your AI doesn't just answer. It acts.
Connect external APIs — check order status, reset passwords, look up accounts — and let the AI invoke them mid-conversation on any surface.
- Define actions once — works across chat, Slack, and voice
- Confirmation prompts for sensitive operations
- Auth headers stay server-side — never exposed to the browser
- Full audit log of every action invoked
- Growth: up to 5 actions · Scale: up to 25
Feed it everything. It remembers everything.
Point Pernoex at your documentation — websites, PDFs, APIs, wikis — and it builds a searchable knowledge base automatically.
Website Crawler
Auto-discovers all pages, handles JS-rendered content, respects robots.txt.
File Upload
Markdown, HTML, PDF, and OpenAPI specs — drag, drop, indexed.
GitHub
Import docs from any repo. Auto-syncs on push.
Confluence
Connect your workspace with one click. Spaces and descendant pages.
llms.txt
Auto-detects LLM-optimized content for cleaner, richer ingestion.
Path restrictions · Crawl control (pause / resume / cancel) · Sitemap seeding · Automatic re-crawl scheduling
From question to code in one click.
Developers don't want to read docs; they want to write code. Pernoex bridges the gap, turning a simple web search into a native IDE integration.
<script
src="https://cdn.pernoex.com/widget.js"
data-api-key="pk_live_5g9..."
></script>See what your customers actually need.
Pernoex doesn't just answer questions — it tells you which questions matter, what's missing from your docs, and how much money it's saving you.
Top Questions
Most-asked questions clustered by topic. See exactly what your users need help with.
Content Gaps
Questions your docs can't answer yet — with AI-suggested drafts to fill them.
Business Impact
Deflected tickets, time saved, money reclaimed — delivered as a monthly report.
Most Explained Text
What users highlight most on your site. Find the confusing spots before they become support tickets.
When AI isn't enough, humans step in — with full context.
Pernoex seamlessly escalates to your support team in Front or Intercom. The agent gets the full conversation, user identity, and page URL. No one starts from scratch.
- Full conversation history transferred to the human agent
- User identity and page URL included automatically
- Supports Front and Intercom out of the box
- Scale plan only
Built-in intelligence you don't have to build.
Train once, deploy everywhere
One knowledge base powers five surfaces — chat widget, inline explainer, Slack, voice, and developer tools. Update your docs once and every surface stays current.
Cited & verifiable
Every response is grounded in your indexed content. Every claim links back to a source. No hallucinations — if it doesn't know, it asks a clarifying question.
Enterprise-ready security
Two-factor authentication, role-based access control, AES-256 encryption at rest, HMAC identity verification, rate limiting, and CSRF protection — built in from day one.
Simple, transparent pricing.
Start free for 7 days on any plan.
Starter
Ship it today. Go live in minutes.
- 4M credits/mo
- 1 project · 1 seat · 1 domain
- The Console
- Embedded & Inline Explainer
- Web crawling
- PDF upload
- Auto-generated llms.txt
- Responds in 50+ languages
- Conversation analytics
- Email support
Growth
Intelligence layer for growing teams.
- 20M credits/mo
- 3 projects · 3 seats · 5 domains
- Console · Embedded · Explainer
- Content gap detection
- MCP access (configurable cap)
- Slack (1 workspace)
- Confluence · GitHub
- Scheduled re-crawls
- Custom Actions (up to 5)
- White-label included
- Voice support — $499/mo (1,500 min included)
- BYOC Voice — $349/mo (bring your own carrier)
- Priority support
Scale
Enterprise control. Automate and integrate.
- 60M credits/mo
- 10 projects · 10 seats · Unlimited domains
- Custom Actions (up to 25)
- Custom Slack bot
- Conversation handoff (Front · Intercom)
- Business impact dashboard
- Analytics export (CSV)
- REST API access
- Webhooks
- Voice support — $1,499/mo (5,000 min included)
- BYOC Voice — $1,049/mo (bring your own carrier)
- Activity audit log
- Dedicated account manager
Starter → Growth
- Voice support — Phone support in 60+ countries, no agents needed
- Slack integration — Answers where your team already works
- MCP / IDE access — Developers query docs straight from their editor
- Custom Actions — Your AI doesn't just answer — it acts
- Content gap detection — See which questions your docs can't answer
- 5× more credits — ~330–1,000 conversations/mo
Growth → Scale
- Conversation handoff — Route to human agents in Front or Intercom
- Business impact dashboard — Measure the ROI your AI layer delivers
- Custom Slack bot — Your brand answers, not ours
- REST API + Webhooks — Build custom integrations with your systems
- Activity audit log — Full compliance trail of every action
- Dedicated account manager — Hands-on support for your team
Enterprise
Custom volume, dedicated support, SLA guarantees, SSO.
How credits work
Credits are consumed when your product thinks. Whether a user highlights text for an inline explanation, asks a question in the console, or queries your docs directly from their IDE via MCP, the cost is based on the complexity of the reasoning pipeline.
- Starter: ~65 to 200 conversations per month
- Growth: ~330 to 1,000 conversations per month
- Scale: ~1,000 to 3,000 conversations per month
If your product needs to think more, you can add a credit pack in one click.
Common questions
Everything you need to know about Pernoex.
Pernoex is a product cognition layer — it gives your product a mind. Instead of a reactive chatbot that waits for users to start a conversation, Pernoex is a product capability. Users can highlight text for inline explanations, ask questions in a chat widget, get answers in Slack threads, talk to your AI over the phone, or query your docs from their IDE. It is a tool, not a ticket deflection widget.
Pernoex includes a fully managed Model Context Protocol (MCP) server. When a developer asks a question on your site, they can click "Continue in IDE" to copy the full context. They can also install your MCP server in VS Code, Cursor, Claude Code, Windsurf, or JetBrains to query your product's knowledge base directly from their editor.
Just give us your website URL. Pernoex automatically crawls your documentation, API references, help centers, and blogs. You can also connect Confluence workspaces, GitHub repos, upload PDFs, and it auto-detects llms.txt for cleaner ingestion.
Every response is grounded strictly in your indexed content. Pernoex uses a sophisticated retrieval pipeline with multi-query fan-out and Reciprocal Rank Fusion to find the right context. Every claim includes verifiable source citations. If it doesn't know, it asks a smart clarifying question instead of guessing.
Yes. Pernoex offers AI voice support with phone numbers in 60+ countries. Customers call a real number and your AI answers using the same knowledge base as your chat widget. It includes live transcripts, call recordings, and can transfer to a human when needed. Voice is available as a Growth or Scale add-on.
Yes — with Custom Actions, you can connect external APIs so the AI can check order status, reset passwords, look up accounts, and more. Actions work across all surfaces (chat, Slack, voice) and include confirmation prompts for sensitive operations. Growth plans get up to 5 actions; Scale plans get up to 25.
Two things: the question is flagged as a content gap so you can fill it, and if conversation handoff is enabled (Scale plan), Pernoex seamlessly transfers the user to your support team in Front or Intercom with the full conversation context. No one starts from scratch.
Yes. Pernoex is designed to feel native. You can customize the theme using 13 properties in the theme configuration, choose between floating, embedded, or inline modes, and the assistant automatically detects which page the user is on to provide context-aware suggestions. The widget renders inside a Shadow DOM boundary so it never conflicts with your existing styles.
Your product already has the answers.
Give it a voice. One script tag. Under five minutes.
Free 7-day trial · Cancel anytime