What does the Context Manager do?

It helps you build, order, and trim the context you send to an LLM so it fits the window and prioritizes what matters.

Yes. It is free and open source, with no login.

Is my content uploaded?

No. Organization and preview run locally in your browser.

Can it export the context?

Yes, as XML or JSON for use in your prompts or pipeline.

Why order chunks by relevance?

Models weight earlier context more and windows are limited, so ordering improves answers and avoids truncation.

Context Manager

Organize and export your LLM context windows

Help

Next steps

Project contexts

Create a new window to organize your context

Select or create a context

Stop hitting token limits. Curate exactly what the AI needs.

Organize your project code, API docs, and instructions into a single AI-optimized payload.

Code ReviewSelect only relevant files, skip configs and tests

Bug FixInclude error log, failing component, related types

Feature BuildBundle schema, patterns, and spec in one context

or drag & drop files here

About this tool

The Context Manager helps you assemble and optimize the context you send to an LLM. Add chunks, order them by relevance, and preview exactly what the model will receive, with intelligent chunking to fit your window. Export the result as XML or JSON for your pipeline.

How to use it

Add your content. Paste or drop the pieces of context you want to include.
Order by relevance. Arrange chunks so the most important context comes first.
Preview the window. See the assembled context and how it fits your token budget.
Export. Copy or export the optimized context as XML or JSON.

Frequently asked questions

What does the Context Manager do?: It helps you build, order, and trim the context you send to an LLM so it fits the window and prioritizes what matters.
Is it free?: Yes. It is free and open source, with no login.
Is my content uploaded?: No. Organization and preview run locally in your browser.
Can it export the context?: Yes, as XML or JSON for use in your prompts or pipeline.
Why order chunks by relevance?: Models weight earlier context more and windows are limited, so ordering improves answers and avoids truncation.