// The platform

The entire AI stack — under your name. The whole AI system — running under your name.

LINAsystems is one engine and twelve integrated modules — a public web chatbot, a 24/7 AI phone agent, a self-correcting knowledge base, TCPA-safe outbound engagement, a supporter intelligence layer, and a full Command Center — running as your own brand on one managed bill. This is not a chatbot bolted onto your site. It's the whole operational stack, grounded in your knowledge, with a fixed safety floor and human escalation built in.

LINAsystems is one system with twelve modules built in — a public web chatbot, an AI phone agent that answers day and night, a help library that fixes its own mistakes, compliant outbound messaging, a layer that tracks supporter signals, and a full Command Center dashboard — all running under your own brand on one bill. This isn't a chatbot stuck onto your site as an afterthought. It's the entire system you need to talk with people, built on your own knowledge, with built-in safety rules and a way to bring in a real person when needed.

Request a Demo See the two editions →

// One engine

One engine, one knowledge base, one contact record One system, one knowledge base, one contact record

Every module reads and writes the same brain. There is no syncing chat to the phone, no exporting from one tool to import into another, no separate vendor coordinating with another vendor. A fact lives in one knowledge base; a person lives in one contact record; an answer is governed by one safety floor. That single-engine design is what makes the platform behave like a single, coherent operation instead of a pile of bolted-on point tools.

Every module reads from and writes to the same memory. Nothing has to be synced between chat and phone, nothing has to be exported from one tool and imported into another, and there's no second vendor to coordinate with. A fact lives in one knowledge base. A person lives in one contact record. Every answer follows the same safety rules. Because it's all one system, it works like one connected operation instead of a pile of separate tools stuck together.

Twelve modules, one engine — every channel reads and writes the same knowledge, the same contact record, the same safety floor. Twelve modules, one system — every channel shares the same knowledge, the same contact record, and the same safety rules.

Correct an answer once in the Command Center and the fix is live in the next message — across chat, phone, and email — the same turn. No retraining, no developer, no redeploy.

Fix a wrong answer once in the Command Center dashboard, and it's corrected in the very next message — on chat, phone, and email, all at the same time. No retraining, no developer needed, no waiting for a new release.

Entitlements are enforced the same way. Modules you haven't bought don't sit behind a hidden button in the UI — every endpoint is server-enforced, so a plan that doesn't include a module returns an HTTP 403, not a quietly-disabled control. What you pay for is exactly what's live, and access control (admin / editor / viewer) is checked on the server for every privileged action, not just in the interface.

What you're allowed to use works the same way. A module you haven't paid for isn't just a hidden button — it's blocked on the server itself, so trying to use it returns a clear "not allowed" error (HTTP 403) instead of a button that quietly does nothing. You get exactly what you paid for, and access levels (admin / editor / viewer) are checked on the server for every sensitive action, not just hidden in the screen design.

One self-correcting knowledge base grounds chat and phone alike — fix it once, it's fixed everywhere.One knowledge base feeds both chat and phone — fix something once, and it's fixed everywhere.
One unified contact record dedupes a person across chat, phone, and email — by email, then phone, then session.One combined contact record for each person, even if they reach you by chat, phone, and email — matched first by email, then phone, then session.
One fixed safety floor — no fabrication, always discloses it's an AI, never reveals its prompt — that operator settings cannot override.One set of safety rules that can't be turned off by an operator — it never makes things up, always admits it's an AI, and never reveals its instructions.
One console controls all of it: facts, prompts, website, knowledge, people, and roles.One dashboard controls everything: your facts, your instructions to the AI, your website, your knowledge base, your contacts, and who on your team can access what.

// 5 channels, 12 modules

Five headline channels. Twelve modules underneath. Five main ways people reach you. Twelve modules underneath.

When you talk about LINAsystems you'll hear two numbers — five headline channels and twelve modules. They describe the same system at two levels of zoom, not a discrepancy. The five channels are how a prospect experiences the platform; the twelve modules are the engineering catalog that delivers those channels, plus the outbound, intelligence, and platform layers that keep them consistent and growing.

You'll hear two numbers about LINAsystems — five and twelve. That's not a mismatch; it's the same system described two ways. The five channels are how a prospect experiences the platform day to day. The twelve modules are the full list of building blocks behind those channels, plus the outbound messaging, supporter tracking, and core pieces that keep everything working together.

5Headline channels — public web chat, a 24/7 AI phone agent, a self-correcting knowledge base, a unified CRM, and a full Command Center.Main channels — web chat, an AI phone agent that's always on, a help library that fixes its own mistakes, one combined contact list, and a full Command Center dashboard.

12Integrated modules — the five channels plus the staff AI, website editor, email ingestion, smart alerts, outbound engagement, supporter intelligence, and the platform core.Modules in total — the five channels above, plus the staff AI writing tool, the website editor, email handling, smart alerts, outbound messaging, supporter tracking, and the core foundation.

1Engine, one managed bill, one dedicated isolated install per customer — your brand, your data, your keys.System, one bill, and your own separate, private setup — your brand, your data, your keys.

// The catalog

The twelve modules, in depth All twelve modules, explained

Every plan is assembled from these twelve modules. The first five map directly to the headline channels and each has a deep-dive page; modules six through ten are covered in detail inside the Command Center; modules eleven and twelve — Outbound Engagement and Supporter Intelligence — each have their own deep-dive pages. Every plan starts from the edition core (Chatbot, Knowledge, Command Center, RBAC, and Notifications); bundles package the common modules together, and you can add any module to the core individually.

Every plan is built from these twelve modules. The first five match the main channels above and each has its own page with more detail; modules six through ten are covered inside the Command Center dashboard; modules eleven and twelve — Outbound Engagement and Supporter Intelligence — each have their own pages too. Every plan starts with the edition's core (Chatbot, Knowledge, Command Center, access roles, and Notifications); bundles group the common modules together, and you can also add any single module to the core on its own.

Public AI Chatbot

The public-facing entry point: a branded web chat assistant that opens with an intake gate capturing name, email, phone and ZIP before answering, so no interested visitor is lost between 'interested' and 'reached a human.' It answers from your knowledge base, escalates uncertain or sensitive questions to your team, and on Campaign installs runs voter-guide, endorsement and donation-inquiry flows. Bilingual out of the box; discloses it's an AI when asked.

The first thing visitors meet: a web chat assistant with your branding that asks for a name, email, phone, and ZIP code before it starts answering, so an interested visitor never gets lost before reaching a real person. It answers using your knowledge base, hands off unclear or sensitive questions to your team, and on Campaign installs can help with voter guides, endorsements, and donation questions. It works in more than one language right out of the box, and it tells the truth if someone asks whether it's an AI.

Public AI Chatbot →

Knowledge Engine

A self-correcting knowledge base that grounds every answer. Operators correct any answer from the Command Center and the fix is live in the next message across chat AND phone — no retraining, no developer, no redeploy. Standing notes are the supreme source of truth, overriding stale data within a fixed safety floor. The knowledge base refreshes from your own site daily, staff email blasts can be promoted through a human review gate, and a knowledge graph tracks entity relationships for deeper retrieval accuracy.

A knowledge base that fixes its own mistakes, and every answer is based on it. Your team can correct any answer from the Command Center dashboard, and the fix shows up in the very next message on both chat and phone — no retraining, no developer, no waiting for a new release. Your written notes always win over outdated information, within the platform's fixed safety rules. The knowledge base updates itself from your own website every day, staff email blasts can be added to it after a quick human check, and it tracks how facts relate to each other for more accurate answers.

Knowledge Engine →

AI Phone Agent

A real voice agent built on a Twilio + Google Gemini Live audio bridge that answers your line 24/7, grounds answers in your knowledge base, and warm-transfers or alerts staff on uncertain or sensitive calls. Every call is recorded and reuses the same escalation engine as chat, so a phone escalation fires the same email and alert. Per-call cost controls — a configurable time limit with a spoken warning, plus concurrency caps — bound the cost of every call.

A real voice agent, built on Twilio and Google Gemini Live, that answers your phone line day and night, bases its answers on your knowledge base, and brings in staff — by transfer or by alert — on calls that are unclear or sensitive. Every call is recorded, and it uses the same hand-off system as chat, so a phone hand-off sends the same email and alert. Cost controls — a time limit with a spoken warning, plus a cap on how many calls run at once — keep every call's cost in check.

AI Phone Agent →

Staff Internal AI (MIND)

An internal staff drafting assistant with its own login and templates that writes in the first person as the organization's principal, turning a rough note into on-voice email, replies, and memos in minutes. Tone and topic are unrestricted, but accuracy-first still applies — it will not fabricate facts. Saves draft templates and recent drafts for the team.

A writing assistant for your staff, with its own login and templates, that writes as if it were your organization's leader speaking directly — turning a rough note into a finished email, reply, or memo in minutes. It can write about any topic in any tone, but it still won't make facts up. It saves templates and recent drafts so the team can reuse them.

Staff Internal AI (MIND) →

Command Center + RBAC

The single operating console for the whole platform, white-labeled down to the admin console itself. Three server-enforced roles — admin, editor, viewer — gate every privileged action: viewers are read-only, editors get everything except account and credential surfaces, admins get full control. Every privileged action is written to an append-only audit trail, login events fire subscribable alerts, and sessions use short-lived secure cookies. A Dispatch board surfaces what needs attention across every surface: escalations, callbacks, intent flags, and send-queue approvals in one count.

The one dashboard that runs the whole platform, carrying your brand all the way down to the admin login screen. Three access levels, checked on the server every time — viewers can only look, editors can do everything except manage accounts and credentials, and admins have full control. Every sensitive action is permanently logged, login attempts can trigger alerts, and sessions use short-lived, secure cookies. A Dispatch board collects everything that needs your attention in one place: hand-offs, callbacks, flagged interest, and messages waiting for approval.

Command Center + RBAC →

AI Website Editor

Edit your live public homepage by typing plain-English instructions. The AI emits precise structured DOM and CSS edits applied server-side with a one-shot self-heal if a selector misses, plus live preview. Saves are protected so a bad edit rolls back to the clean baseline.

Change your live public homepage just by typing what you want in plain English. The AI makes precise edits on the server, and if something doesn't quite land, it automatically tries again to fix it, with a live preview so you can see the change first. Saves are protected, so a bad edit can always be rolled back to a clean version.

In the Command Center →

Email Ingestion

Routes inbound staff email (via Microsoft Power Automate / Outlook 365 sync and Resend receiving), classifies blasts versus threads, canonicalizes staff addresses, and auto-promotes marketing blasts into knowledge while threaded conversations pass through a human review queue before publishing. Notes marked internal or do-not-touch are excluded end to end.

Sorts incoming staff email (synced through Microsoft Power Automate / Outlook 365, and through Resend), tells mass marketing emails apart from back-and-forth conversations, cleans up staff addresses, and automatically adds marketing emails to the knowledge base while real conversations go through a human check first before publishing. Anything marked internal or "do not touch" is kept out of the knowledge base entirely.

In the Command Center →

Smart Alerts

Operator-subscribable notifications across email and optional Telegram push, with configurable recipients, quiet hours, and custom triggers. Alerts fire on escalations, login events, deploys, and knowledge/health events. Custom signals let operators bind any internal system event to a self-service notification without writing code. Delivery is best-effort — a Telegram outage never breaks a live request.

Notifications your team can subscribe to, sent by email and optionally by Telegram, with control over who gets them, quiet hours, and what triggers them. Alerts go out for hand-offs, logins, software updates, and knowledge or system-health events. Your team can also set up their own custom alerts without writing any code. Delivery never blocks a live request — if Telegram is down, nothing else breaks.

In the Command Center →

Unified CRM

A single contact database that merges chat, phone, and email interactions into one record per person, auto-deduplicated by email then phone then session. One-click CSV export (donor/contact exports gated and audit-logged), aggregated first/last-seen and session counts, and permanent per-person erase that scrubs PII across all channels. Sensitive PII is redacted by default in admin views. Integrates with the supporter intelligence and outbound engagement modules for a full-lifecycle view.

One contact list that combines chat, phone, and email into a single record per person, automatically matched by email, then phone, then session, so no one shows up twice. You can export contacts to a spreadsheet with one click (donor and contact exports are restricted and logged), see when each person first and last showed up and how often, and permanently delete a person's personal information across every channel on request. Sensitive personal details are hidden by default in admin views. It connects with the supporter tracking and outbound messaging modules for a complete picture of each person.

In the Command Center →

Platform Core

The shared foundation every plan includes: a config and credentials registry, the immutable prompt safety floor, crash-safe atomic state writes, rate limiting, and the model-override chain that lets an admin switch the live LLM (Google Gemini or OpenAI) from the dashboard with no redeploy. Includes a read-only Public API v1 (API-key gated, rate-limited, default-off) and an outbound webhooks system for integrating the platform with external tools. Secrets live in a managed secret store and are never displayed, logged, or returned by the API.

The shared foundation that every plan includes: a registry for settings and credentials, the safety rules that can never be changed, crash-safe data saving, limits on request rates, and the ability for an admin to switch the AI behind it (Google Gemini or OpenAI) from the dashboard with no need to redeploy anything. It includes a read-only public API (locked behind a key, rate-limited, off by default) and a way to send outbound webhooks to connect with other tools. Passwords and keys are stored in a secure vault and are never shown, logged, or returned by any API.

In the Command Center →

Outbound Engagement

A TCPA-safe outbound messaging layer for SMS and email, built on a six-gate send queue: feature flag on, rail master on, human approval, under the daily cap, not quiet hours, recipient not suppressed — all six must hold before any message leaves. Double-send-safe architecture. Includes Twilio-backed outbound SMS with automatic STOP opt-out compliance, an automated followup producer, and a full delivery log. Default-off; operators open gates deliberately.

A legally compliant way to send outbound text messages and email, with six separate checks that must all pass before any message goes out: the feature must be turned on, the channel must be turned on, a human must approve it, it must be under the daily limit, it must not be during quiet hours, and the recipient must not have opted out. It's built so a message can never accidentally send twice. It includes text messaging through Twilio with automatic handling of STOP opt-outs, a tool that drafts automated follow-ups, and a full log of everything sent. It's off by default — your team has to turn each gate on deliberately.

Outbound Engagement →

Supporter Intelligence

A passive sensing and analytics layer that classifies every chat and phone turn for intent and sentiment at conversation time — no message text stored, no latency added. Signals accumulate into enriched supporter profiles, get distilled into durable persona segments (high-intent, multi-channel, at-risk, repeat caller), roll up into channel attribution analytics, and power deterministic A/B experiments. The Dispatch board aggregates what needs a human across all surfaces. Experiments are default-off.

A background tracking layer that reads the intent and tone of every chat and phone exchange as it happens — without storing the actual message text, and without slowing anything down. These signals build up into richer supporter profiles, get grouped into lasting categories (highly interested, reaches you multiple ways, at risk of disengaging, repeat caller), feed into reporting on which channels work best, and power consistent A/B tests. The Dispatch board gathers everything that needs a human's attention into one place. Testing features are off by default.

Supporter Intelligence →

// Go deeper

Explore each module

Public AI Chatbot

The branded web assistant that captures the visitor and carries them through chat, escalation, and donation flows.

The branded web assistant that gets the visitor's info and guides them through chat, hand-offs to a person, and donation questions.

→

AI Phone Agent

An always-on AI front desk that answers 24/7, escalates to staff live, and records every call.

An AI front desk that's always on, answers day and night, brings in staff right away when needed, and records every call.

→

Staff Internal AI (MIND)

Turns a rough note into on-voice email and memos in minutes — your organization's first-person voice, enforced.

Turns a quick, rough note into a finished email or memo in minutes — written to sound like your organization, every time.

→

Knowledge Engine

The self-correcting brain — corrections go live in the next message; standing notes override stale data.

The memory that fixes its own mistakes — corrections show up in the very next message, and your written notes always win over outdated information.

→

Command Center

The white-labeled multi-operator console: RBAC, audit trail, alerts, CRM, website editor, and the platform core.

The single dashboard, carrying your brand, where your team manages access roles, logs, alerts, contacts, the website editor, and the core system.

→

Outbound Engagement

TCPA-safe outbound SMS and email on a six-gate queue — human-approval required, suppression-aware, double-send-safe.

Legally compliant outbound texts and emails, with six checks that must all pass — a human has to approve it, it respects opt-outs, and it never sends the same message twice.

→

Supporter Intelligence

Per-turn intent signals, enriched profiles, persona segments, attribution analytics, and A/B experiments.

Reads intent in every exchange, builds richer supporter profiles, groups people into categories, reports on what's working, and runs A/B tests.

→

Running in production

Built on a platform running in production — the same engine running in a real, live operation, not a prototype.

2,600+ tests passing

The full automated suite runs — and must pass — before every deployment ships.

Live in days, not weeks

About 90% of a rebrand is no-code content set in the dashboard; only ~10% is a small one-time vendor setup.

// Live demo

See it answering real questions right now Watch it answer real questions, right now

The live demo runs the exact system that ships to customers — try it as a campaign or as a business.

This demo is the exact same system real customers get — nothing watered down. Try it set up as a campaign or as a business.

Request a Demo See the campaign demo → See the business demo →