What are Exabase Workers?
Workers are autonomous agents that run inside your Bases on a schedule. They organize, enrich, and maintain your content automatically. Think of them as an agent-maintained wiki. Your knowledge base compounds over time instead of rotting.
Define a task in natural language, set a schedule, and Workers handle it. They run on Exabase's infrastructure. No cron jobs, no servers, no workflow builders.
Problem
Knowledge-bases that rot
Your agent's knowledge base is useful for a week.
Then links go stale, notes pile up untagged, and structure collapses. The more you add, the harder it gets to find anything.
Knowledge decays unless someone maintains it. That someone is usually you.
Solution
Self-updating knowledge-bases
Workers are autonomous agents that maintain your Bases continuously. They research, organize, enrich, summarize, and prune, all on autopilot.
Your knowledge base gets better over time, not worse.
Every new document is automatically tagged, summarized, and placed where it belongs. Every stale link is flagged. The more your agents use the workspace, the more valuable it becomes.
How it works
Create a Worker
Give it a name, a prompt describing what to do, and the Base it runs in.
Using the SDK:
Using cURL:
Set a schedule
Run daily, weekly, monthly, or on a custom cadence. Workers execute on Exabase's infrastructure. Nothing to host or manage.
Or trigger on demand
Run a Worker manually via the API whenever you need it.
Inspect the results
Every run creates a chat session. View the Worker's full reasoning, tool calls, and output in the Exabase console or via the API.
Example Worker capabilities
Search the web
Find and bookmark new content on a schedule. Keep your knowledge base current.
Enrich content
Add metadata, summaries, or context to existing resources. Turn raw storage into useful knowledge.
Create and update resources
Add notes, save links, upload files. Workers produce content, not just move it.
Organize
Move resources into folders, apply tags, rename files. Maintain structure as your workspace grows.
Manage memories
Create, update, and resolve memories as facts change.
Manage tasks
Create and update tasks based on what Workers find or what needs attention.
Powerful access controls
Use cases
Why Exabase
Part of a full platform
FAQs
What is Exabase?
Exabase is infrastructure for AI agents. It gives your agents memory, versioned file storage, AI deep search, and context automation through a set of APIs. Store what your agent learns, search inside any content type, and keep knowledge bases current automatically. Built for production use.
Who uses Exabase?
Developers and teams building AI agents, copilots, and RAG applications. If your agent needs to remember things between sessions, store and retrieve files, search across documents and media, or stay up to date without manual maintenance, Exabase handles that infrastructure so you can focus on your product.
What are Exabase Workers?
Workers are autonomous agents that run tasks inside your Bases on a schedule. You define a task in natural language, and the Worker executes automatically using built-in tools like search, content creation, task management, and web research. They run on Exabase's infrastructure. The result is a self-organizing knowledge base that compounds over time instead of going stale.
How do I automate knowledge base updates?
Create a Worker with a prompt like "Check all bookmarks for broken links, tag any new untagged resources, and organize recent uploads into the right folders." Set it to run daily. The Worker handles maintenance automatically.
Can I run an AI agent on a schedule?
Yes. Workers support daily, weekly, and monthly schedules. Set the hour and timezone. You can also trigger a run on demand via the API at any time.
What tools do Workers have access to?
17+ built-in tools: search resources, read content, create and edit notepads, create folders, rename and move resources, delete resources, apply tags, search tags, create/edit/delete/list tasks, read/write scratchpad, and web search. All available by default.
Can I restrict which tools a Worker can use?
Yes. Pass toolPermissions when creating a Worker. Only the tools you list will be available. Everything else is blocked. Useful for limiting a Worker to read-only operations or preventing it from deleting content.
Can I scope a Worker to specific resources?
Yes. Use focusResourceIds to guide the Worker toward specific resources as context, or accessResourceIds to strictly limit which resources it can access. This prevents the Worker from seeing or modifying things it shouldn't.
How do I see what a Worker did?
Every run creates a chat session. You can view the full conversation through the Exabase console or the API, including the Worker's reasoning, every tool call it made, and its output.
Do Workers run on my infrastructure?
No. Workers run entirely on Exabase's infrastructure. Nothing to deploy, host, or maintain.
Can Workers search the web?
Yes. The web-search tool lets Workers search the web and get results with snippets. Use this for research, competitive monitoring, news aggregation, or enriching your knowledge base with external information.
Can Workers create and edit content?
Yes. Workers can create notepads, edit notepad content, create folders, rename and move resources, and apply tags. They produce content, not just move it around.
How do schedules work?
Set frequency (daily, weekly, monthly), hour (0-23), and optionally timezone. Weekly and monthly workers also need a day value. Activate or deactivate a schedule at any time without deleting the Worker.
What's the difference between Workers and a cron job?
A cron job runs a script on a timer. You write the code, handle errors, manage infrastructure. A Worker is a natural-language agent with built-in tools and full access to your Base. You write a prompt, not a script. It understands content and makes decisions based on context.
What's the difference between Workers and Extract?
Extract processes a specific file or URL and returns structured data. Workers are ongoing agents that perform recurring tasks across your entire Base. You might use Extract to process a new document, and a Worker to monitor for new documents, extract them, summarize them, and organize them automatically.
Can Workers manage tasks?
Yes. Workers can create, edit, delete, and list tasks. Use them to triage incoming tasks, update priorities, flag overdue items, or create follow-up tasks based on what they find in your workspace or Base.
How is this different from Make or other workflow automation tools?
Tools like Make connect apps with triggers and actions. They move data between services but don't understand content. Workers operate inside your data with full context. They can read a document, understand it, and decide how to organize it based on what else is in the workspace. That's not a trigger-action workflow.
What does "compounding knowledge" mean?
Every time a Worker runs, it makes your Base more useful. New documents get summarized and tagged. Stale content gets flagged. Web research gets saved and organized. Over weeks and months, your knowledge base grows in value automatically. Without Workers, knowledge decays. With Workers, it compounds.
Deciding?
Ask your favourite AI about Exabase:
