Zygnal
Now in beta — request early access

Signal in the noise. From the first customer question to the merged fix.

Zygnal is an AI-native platform for support and engineering — living help docs, a learned support memory, tickets that link to your code, and an in-product drawer. One place where customer questions, support work, and engineering fixes all see each other.

Every resolved ticket becomes a captured insight. Insights become new articles and engineering fixes. Engineers triage tickets with code in context — soon, from inside their IDE. The loop closes.

See how it works

Hosted in Sydney (ap-southeast-2). Multi-tenant from day one. Plugged into your code, not just your help desk.

help.your-product.com
How do I reset my password?
AI answer 3 sources

Click Profile → Security, then Reset password. You'll get an email link that lasts 30 minutes.

Reset password guide Account security FAQ
Still stuck? Open a ticket — your engineers see context automatically.
The problem

Support tooling makes you choose. Zygnal does not.

Most teams duct-tape five tools together: a help-doc CMS, a ticket system, an AI chat add-on, a wiki for "things we have learned," and a project tracker the engineers actually open. Each one is partly aware of the others. None of them close the loop, and none of them know about your code.

Help docs go stale

You wrote them six months ago. The product changed twice. Now they're wrong, and customers find the wrong answers before they find your team.

Hard-won knowledge gets lost

Every resolved ticket teaches your team something. Most of that learning lives in someone's head, in Slack threads, or in a Jira comment nobody will read again.

AI without context is useless

Generic AI chatbots start cold every time. They do not know your product, your customers, or what your team has already worked out. So they hallucinate, or punt to human.

The fix loop never closes

Customer reports a problem. Engineer fixes it. Six weeks later, same question from a different customer. Nobody updated the docs. Nobody told marketing. The cycle repeats.

How it works

Four parts, one loop.

Your customers get a helpful AI right inside your product. Your support team gets tickets routed and triaged with full history. Your engineering team gets the ticket, the linked code, and the knowledge fed back as fixes and docs — soon, from inside their IDE.

01

Help docs that stay current

Markdown sources, AI-polished, continuously refreshed. Edit by hand or by source — Zygnal keeps a diff against the original. Each article carries page routes, freshness scores, and a "last validated" date so you always know what is fresh.

02

A learned support memory

Every resolved ticket becomes a structured insight: symptom, cause, resolution. Insights cluster into patterns. Patterns become new help articles and product-improvement candidates. The knowledge grows automatically.

03

Tickets, not a black box

First-class tickets with severity, priority, SLA, sprint backlog, and GitHub branch + PR linking. Threaded replies. Email-to-ticket via Mailgun. Sync with Jira. The canonical record lives with you, not in a vendor silo.

04

An in-product drawer

Embeddable widget your customers see inside your SaaS app. Contextual help, AI chat grounded in your knowledge, ticket creation, file uploads with vision OCR. HMAC-signed identity — no extra signup.

The closed loop:

Customer asks → drawer answers from KB → if not, opens ticket → AI triages with code from the linked repo → engineer ships a fix as a PR → merge updates the ticket, drafts the article, closes the loop with the customer → next customer finds it themselves.

For engineers

The support platform that talks to your code.

Most "support tooling" treats the engineering hand-off as someone else's problem. Zygnal treats it as the point. Tickets link to repos. AI reads the code. Your IDE sees the prep. Fixes feed the knowledge base.

Or, more bluntly: prep is the bottleneck, not coding speed. Zygnal closes the prep gap.

Live in beta
  • Repository linking, scoped per product

    Connect a GitHub repository to a product. Code access stays scoped to that product's repos — your support team can't accidentally read code they shouldn't.

  • Branch + PR linking on every ticket

    One click creates a branch from your default base. PRs that reference the ticket update its status automatically via webhook. The link survives re-titling, re-assignment, and merging.

  • In-ticket AI with code in context

    When triaging a ticket, the AI can read files, search code, and walk the directory tree of the linked repo — surfaced inline as the conversation happens. Every read is audited.

  • Sprints, severity, SLA

    First-class sprint backlog with drag-to-rank, per-product calendars, and breach tracking. Engineering capacity sits next to support load, not in a separate Jira your team forgets to open.

Coming
  • Zygnal MCP server

    A Model Context Protocol bridge that exposes ticket context, KB search, past resolutions, and code search to your IDE. Open Cursor, Claude Code, or Windsurf — your AI sees the full ticket prep without you copy-pasting context.

  • Code indexing for semantic search

    Embeddings over your linked repos so the AI can find relevant code by intent, not just keyword. Becomes another retrieval surface alongside articles and resolution insights.

  • Closed-loop on PR merge

    When the fix lands, Zygnal drafts the resolution insight, the article update, and the customer reply automatically. You review and ship.

Engineer's loop:

Customer reports a bug → support triages with AI, attaches the relevant code → engineer opens the ticket in their IDE via MCP → fix goes out as a PR → merge updates the ticket, drafts the article, closes the loop with the customer.

Features

Everything support, nothing extra to wire up.

The features below are live in the platform today, in active internal use across our own product suite.

Hybrid retrieval

Full-text + semantic search, weighted and tunable. Markdown-aware chunking preserves heading context. Per-tenant + per-product partitioning — no cross-product bleed.

Insight capture & cluster mining

Every resolved ticket is summarised into a structured insight. Insights are clustered to surface emerging patterns and seed new article drafts.

Anomaly detection

Isolation-forest models watch ticket signals — volume spikes, unusual subject clusters, severity shifts. You see issues before they snowball.

Sprints & GitHub linking

Tickets sit in sprints. Engineers link a branch and PR with one click. Resolution closes the ticket and feeds the insight pipeline automatically.

GitHub App + in-ticket code access

Connect a repo per product. The triage AI can read files, search code, and walk the directory tree — scoped to that product's repos and audited end-to-end. MCP server for Cursor / Claude Code / Windsurf is coming.

Federation across products

Pull articles from existing help systems for unified search. Phase-1 customers federate with their existing knowledge base; nothing has to migrate on day one.

Email-to-ticket

Inbound email parsed and threaded as tickets via Mailgun. Replies from inside Zygnal go back out as email. The customer never sees a portal change.

Per-tenant SLA engine

Business calendars per tenant. Override profiles per product or per customer. Breach tracking built in. SLA is a first-class citizen, not a stitch-on report.

Audit log + multi-tenancy

Every mutating action is audit-logged via middleware. Tenant + product isolation enforced at the query layer, with row-level security in phase 2.

In-product drawer

HMAC-signed boot, JWT-scoped chat, contextual help by page route, ticket creation, file uploads with Vision OCR. Zero extra signup for your customers.

Pricing

Per product, per month. No per-seat charge.

Drawer MAU per product is the value axis. Unlimited staff seats on every tier. 30-day free trial. Annual prepay saves 20%.

Phase 2 launch pricing. Zygnal is in beta and not yet billed externally. to lock in beta terms when we open up.

Base Zygnal — per product per month

Starter

$499 /product/mo

Up to 2,500 drawer MAU

Core platform: KB articles, tickets, drawer with local AI chat, sprints, GitHub linking, federation, insight capture, full admin.

  • Unlimited staff seats and end-customers
  • Unlimited articles, tickets, insights, sprints
  • Help portal with per-product branding
  • Drawer with local AI chat (RAG-grounded)
  • Email-to-ticket via Mailgun
  • Jira import + webhook
  • GitHub branch + PR linking
  • Federation across products
  • Email support
Most popular

Growth

$1,499 /product/mo

Up to 10,000 drawer MAU

Everything in Starter, plus the closed-loop intelligence: cluster mining, improvement drafts, anomaly detection.

  • Everything in Starter
  • Cluster mining + improvement drafts
  • Anomaly detection on ticket signals
  • Higher AI allowance
  • Priority support

Scale

$3,999 /product/mo

Up to 30,000 drawer MAU

For teams that need enterprise controls: SSO/SCIM, customer-managed keys, audit-log streaming, multi-region.

  • Everything in Growth
  • SSO / SCIM (Entra, Google, SAML)
  • Per-tenant KMS keys
  • Custom SLA
  • Audit-log streaming
  • Multi-region option
  • Dedicated support

Zygnal Agent — add-on per product per month

Upgrade drawer chat from KB Q&A to a real support agent (Clara). Multi-step reasoning, search past resolutions, look up customer data, take actions across systems. Powered by KernelService under a white-label arrangement.

Starter Agent

+$499 /product/mo

Up to 500 chat-active MAU

Growth Agent

+$1,499 /product/mo

Up to 2,000 chat-active MAU

Scale Agent

+$4,499 /product/mo

Up to 6,000 chat-active MAU

Annual prepay

20% off list price.

Multi-product discount

2–3 products: −10%. 4+ products: −20%. Same tenant.

Free trial

30 days, single product, all features except SSO/KMS, 1k MAU cap. No card upfront.

Trust

Built for B2B SaaS, hosted where you'd expect.

Zygnal is being built for our own product suite first. The same posture you'd want for your customers' data is the posture we hold ourselves to.

Australian data residency

Hosted in AWS ap-southeast-2 (Sydney). Data does not leave the region by default. Multi-region available on Scale.

Multi-tenant from day one

Tenant + product isolation enforced at the query layer. Phase 2 adds Postgres row-level security and per-tenant KMS keys for Scale.

Audit log on every mutation

Middleware-driven audit logging on all writes. Stream to your SIEM on Scale. Nothing happens in the system that is not recorded.

HMAC-signed drawer identity

The in-product drawer authenticates customers via HMAC-signed boot tokens issued by your backend. Your drawer, your identity, no extra signup.

SOC 2 on the roadmap

SOC 2 Type II is a phase-2 commitment, scheduled alongside the commercial launch. Compliance posture is being built deliberately, not retrofitted.

Vendor-independent ticket data

Tickets and resolutions are first-class records in your tenant. Federation is one-way pull from external sources. You can leave; your knowledge comes with you.

FAQ

Common questions.

Is Zygnal generally available?
Not yet. Zygnal is in beta, currently used internally across the team’s own product suite. We are taking early-access requests now and will open up self-serve signup in phase 2 once internal use validates the value.
What is a "drawer MAU" and what is a "chat-active MAU"?
Drawer MAU is a unique end-customer who opens the in-product drawer at least once in a calendar month, scoped to your tenant + product. Chat-active MAU is the subset who actually starts a chat session — opening and bouncing does not count. Both are visible live in the dashboard.
How does the in-product drawer authenticate users?
Your backend issues an HMAC-signed boot token containing the customer identity. The drawer presents that token; Zygnal verifies the signature and scopes the session. There is no separate signup, no shared secrets in the browser, and no Zygnal account required for end-customers.
How does Zygnal work with our existing Jira / GitHub / help system?
Tickets sync with Jira via webhook + import. GitHub branches and PRs link directly to tickets. Existing help articles can be federated in (one-way pull) for unified search before you migrate any content. Nothing has to move on day one.
Is this just for support teams, or also for engineers?
Both. Today, engineers get one-click branch creation from tickets, PR-state webhook sync, and an in-ticket AI that can read files and search code from the linked repo while triaging. Coming next: a Zygnal MCP server so Claude Code, Cursor, and Windsurf can pull the full ticket context — KB articles, past resolutions, and code search — directly into the editor. Prep is the bottleneck, not coding speed.
How does the AI access code without leaking it?
Code access goes through our GitHub App, scoped per product to the repos your team explicitly links. The triage AI uses three tools: read a file at a ref, search code, and list a directory — every call is recorded in an audit log with who, what, and why. No code is stored in the AI provider or used for training.
Where is data hosted?
AWS ap-southeast-2 (Sydney). Data does not leave the region by default. Scale tier customers can opt into multi-region deployment. Per-tenant customer-managed KMS keys are available on Scale.
How is AI used? Is my data sent to third-party LLMs?
AI is used for chunk summarisation, embedding generation, and drawer chat. We use Anthropic and OpenAI under enterprise zero-retention agreements. Your content is not used to train external models. The agent layer (Zygnal Agent add-on) is powered internally by KernelService — invisible to your customers.
What is the difference between Zygnal and Zygnal Agent?
Zygnal (base) gives you a drawer with grounded KB chat — answers from your help articles, with sources. Zygnal Agent upgrades that drawer to a multi-step agent (Clara): it can search past resolutions, look up customer data, and take actions across systems. You can subscribe to base only, then add the agent later.
Can our engineers self-host?
Not in phase 1. Enterprise self-hosting may be considered case-by-case in phase 2 as part of a Scale-tier conversation, but the product is built and operated as a hosted SaaS.
How do you handle SLAs for your customers?
Per-tier SLAs published with the contract. Internally, Zygnal has a per-tenant SLA engine — business calendars, override profiles per product or customer, breach tracking. We use it on ourselves first; what we promise externally is what we operate against internally.
How do I get early access?
Request early access from any "Get early access" button on the site. We are gradually opening Zygnal to external customers as internal use stabilises. Early-access customers get priority onboarding and locked-in beta pricing.
Get on the list

Be one of the first.

Zygnal is opening up to early-access customers as internal use stabilises. Tell us about your product and we'll be in touch when there's a slot.

Talk to us