Browse the docs
Documentation
Learn ikazebot, step by step.
The user guide takes you from an empty workspace to an agent answering your customers. The developer guide explains how the product is built and how to work on it.
Start here
Read the introduction- Your account and workspaceSign up, verify your email, create a workspace and switch between workspaces.
- Create your first agentName it, pick a model, write its instructions and set its temperature.
- Knowledge sourcesTeach an agent from files, websites, text and questions with their answers.
- Test in the playgroundTry changes without saving, compare two models and inspect exactly what a reply was given.
- Website widget and help pagePut an agent on your site with one script tag, style it and set its guardrails.
Need help?
User guide
For businesses and their teams: build an agent, train it, put it in front of customers and run it day to day.
Getting started
- IntroductionWhat ikazebot is, the five ideas everything else builds on, and how a reply is made.
- Your account and workspaceSign up, verify your email, create a workspace and switch between workspaces.
- Create your first agentName it, pick a model, write its instructions and set its temperature.
- Build an agent with AIDescribe your business and answer a few questions; the agent is built from your answers.
- Test in the playgroundTry changes without saving, compare two models and inspect exactly what a reply was given.
Build and train
Deploy and channels
- Website widget and help pagePut an agent on your site with one script tag, style it and set its guardrails.
- Widget JavaScript and identityControl the widget from your page, listen to its events and verify signed-in users.
- WhatsApp, Messenger and InstagramConnect your Meta app so an agent answers customers in their messaging apps.
- REST APIUse agents, conversations and contacts from your own systems with an API key.
- WebhooksHear about conversations as they happen, and verify that each call came from us.
Operate and manage
- Inbox and human takeoverWatch conversations live, take one over, reply, write internal notes and hand back.
- HelpdeskTags, priorities, statuses, fields, teams, triggers, business hours, macros and ratings.
- Contacts and leadsYour customers, the attributes you keep about them, and the leads agents collect.
- AnalyticsChats, visitors, feedback, topics and sentiment, compared with the period before.
- Team and rolesInvite people and choose what each of them may do.
- Plans, credits and billingHow credits are spent, where to see usage, and how to buy a plan or more credits.
- Privacy and your dataRetention, exports, and deleting your account.
Developer guide
For engineers working on ikazebot itself: how it is built, how to run it, and the rules that keep tenants apart.
Foundations
Data and tenancy
Server
- API conventionsConfiguration, the error contract, mirrored contracts and authorisation.
- AI runtimeHow a reply is assembled, retrieved, streamed and charged.
- Knowledge pipelineIngestion, chunking, embeddings and the SSRF-safe crawler.
- Background jobsBullMQ queues, the worker process, retries and dead letters.
Web app
Shipping
- TestingThe gate, and the tenant-isolation suites that block a merge.
- DeployingThe Dokploy runbook: services, environment, migrations and order of deploys.
- Platform administrationBecoming a platform admin, curating models, approving payments and watching jobs.
- Writing documentationWhere these pages live, how to add one, the Markdown they support and the checks they pass.