Skip to content

docs: clarify the README opening around the self-improvement loop #1

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
lablnet merged 1 commit into from
Apr 15, 2026
+14 −4
Merged

docs: clarify the README opening around the self-improvement loop #1

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 14 additions & 4 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,21 +3,31 @@
<img src="assets/logo.png" width="150" alt="HyperFlow Logo" />
</p>

> **⚠️ EXPERIMENTAL**: This project is currently in an experimental phase and is **not recommended for production use**.
HyperFlow is a framework for building agents that rewrite and test their own improvements.

Instead of manually retuning prompts and logic after every failure, HyperFlow runs a self-improvement loop where an agent evaluates what happened, edits its own code, tools, and prompts, then tests the new version in a sandbox.

The core idea is simple: do not just rerun the same workflow. Learn from execution and get better over time.

Self-improving agent framework powered by LangChain and LangGraph.
Built on LangChain and LangGraph. Inspired by [HyperAgents](https://github.com/facebookresearch/HyperAgents) (Meta Research, 2026).

Inspired by [HyperAgents](https://github.com/facebookresearch/HyperAgents) (Meta Research, 2026).
> **⚠️ EXPERIMENTAL**: This project is currently in an experimental phase and is **not recommended for production use**.

## What it does

HyperFlow runs an evolutionary self-improvement loop where a **MetaAgent** rewrites its own source code (including prompts, tools, and logic) to make it better at solving tasks. It is **self-referential**: the mechanism that improves the agent is itself part of the editable code. Each generation:
HyperFlow runs an evolutionary self-improvement loop with two roles:
- **TaskAgent** solves the domain problem
- **MetaAgent** studies evaluation results and improves the system

Each generation:

1. Select a parent generation from the archive
2. MetaAgent reads past evaluation scores and edits the source code
3. Evaluation scripts run in a sandbox to score the new agent
4. Better agents are added back to the archive for future generations

It is **self-referential**: the mechanism that improves the agent is itself part of the editable code.

> [!IMPORTANT]
> This framework is currently in an **Experimental** state. See [Limitations](docs/limitations.md) for more information.

Expand Down