Skip to main content
Every error an Ultron agent encounters becomes a lesson. Every lesson is recalled before the next run. Errors happen at most once.

​
How it works

1

Error occurs

An agent encounters a failure: an API rate limit, a scraping block, a tool returning an unexpected format, a cold email failing the quality gate.
2

Lesson stored

The agent immediately stores a feedback memory:
3

Next run starts

Before starting any work, the agent searches memory for lessons tagged with its name:
4

Lessons applied

The retrieved lessons are injected before any tool calls. The agent adjusts its behavior — batching correctly, using alternative approaches, avoiding known failure patterns.

​
Real examples

​
Why both errors AND successes

The lessons system stores confirmed patterns too:
Storing only failures causes drift. Ultron starts changing what’s working, trying to improve. Storing validations locks in what works.
You can manually add lessons. In chat, say: “Remember that [thing you learned] — apply this going forward.” Ultron stores it as a feedback memory immediately.