Skip to main content
AI Agents are AI assistants you configure inside Bold to handle the administrative and repetitive tasks that eat into your team’s day. Instead of writing scripts or building integrations, you simply tell each agent what to do in plain language — the same way you would explain a task to a new colleague. The agent then queries your Bold data, takes action, and runs whenever you decide, freeing your people to focus on keeping the factory flowing.

​
How Agents Differ from Traditional Automation

Classic automation follows rigid, pre-programmed steps. If the input changes — a customer emails in a different format, a shift ends late — the script breaks and someone has to fix it manually. Bold AI Agents work differently:

Understand Context

Agents read and reason about your actual Bold data rather than pattern-matching fixed inputs.

Adapt to Change

When inputs vary, agents adjust their approach without requiring you to reconfigure anything.

No Code Required

Instructions are written in natural language — any manager can create and adjust agents independently.

​
Key Capabilities

Every agent you create in Bold is defined by three configuration elements:
1

Natural-Language Instructions

Describe the task exactly as you would explain it to a team member. The agent understands intent, not just keywords, so you don’t need to learn any special syntax.
2

Scoped Permissions

You decide precisely which modules the agent can read, and which records it can create or edit. An agent never touches anything outside the boundaries you set.
3

Custom Scheduling

Choose whether the agent runs on demand — when you or a colleague talk to it — or on a recurring schedule at whatever frequency your factory needs.

​
What an Agent Can Do

Anything a team member currently does inside Bold is a candidate for delegation. Agents can:
  • Query Bold data — production records, inventory levels, operator timesheets, maintenance logs.
  • Create and edit records — new orders, updated recipes, corrected timesheet entries.
  • Generate and send reports — formatted summaries delivered by email on a schedule you set.
  • Close processes — pause open operations, mark tasks complete, finalise shift data.
  • Send emails — notify colleagues, suppliers, or customers based on conditions in your data.
If a task is repetitive and lives inside Bold, an agent can almost certainly handle it. If you have a use case in mind, talk to your Account Manager — they will help you build the first agent and get you up to speed quickly.

​
AI Credits: How Consumption Works

Bold measures agent work in AI Credits. Every time an agent runs, it consumes credits based on:
  • How much data it needs to query
  • How much reasoning it has to perform
  • How many actions it ultimately takes
As a practical reference, each “click” a person would make inside Bold to query or change something is roughly equivalent to 1–3 credits. You also choose an intelligence level for each agent:
Intelligence LevelBest ForCredit Usage
Lighter modelSimple, well-defined tasksLower
Powerful modelComplex reasoning, ambiguous inputsHigher
You can monitor each agent’s credit consumption at any time from within Bold to understand which agents are delivering the most value.

​
Plans and Credit Allocations

Lite

5,000 credits / monthIdeal for teams getting started with a handful of automated tasks.

Pro

20,000 credits / monthFor factories running multiple agents across departments.

Enterprise

By arrangementVolume-adjusted to match your operational scale.
If you run low on credits before the end of the month, you can top up by speaking with your Account Manager. Top-up packs are available on all plans.