Skip to main content

Extending and Contributing to DeerFlow

1. Start with the workflow stage you want to improve​

DeerFlow contributions are easiest when you pick one clear layer:

GoalFirst place to look
Change orchestration logicsrc/ runtime and graph logic
Improve the web experienceweb/
Add examples or patternsexamples/
Clarify operationsdocs and harness guidance

2. Why this matters​

Because DeerFlow is step-oriented, it is especially important not to mix planning, evidence gathering, synthesis, and UI changes into one large patch unless they truly belong together.

3. Best first contributions​

The easiest high-value first contributions are usually:

  1. docs clarifications,
  2. example workflows,
  3. better prompts or stage boundaries,
  4. small integration fixes,
  5. output-quality improvements.

4. Think in reusable workflows​

A strong DeerFlow contribution usually makes one of these things better:

  • research reliability,
  • workflow clarity,
  • human review ergonomics,
  • report quality.

5. Before opening a PR​

Validate the full research flow your change touches. A patch that looks correct in one function but breaks the end-to-end research process is still a real regression.