Free · no signup · browser-local
Agent Context Window Budget Planner
Plan a user-defined context window with labeled text segments, optional content, output reserve and manual overhead. Inspect remaining capacity and export the allocation.
Local workspace · no model calls · inputs are not sent or saved
Prepare your text
Exact text encoding for the selected vocabulary. Chat wrappers, tool definitions, images, reasoning and actual API billing are outside this count. Special-token-looking strings are ordinary text.
Defaults are an illustrative allocation, not a model specification. Set your own limit. Each enabled segment is encoded separately; labels do not count. Up to 50 segments and 100,000 total UTF-8 bytes, including disabled text.
Run again after editing. Calculations do not run on each keystroke.
Your output
agent-context-budget.jsonComplete the input to preview and export your result.
Getting started
Set your own window and reserves, add labeled segments, enable the content to include, then calculate and export JSON or Markdown.
Before you use the result
Up to 50 segments and 100,000 total UTF-8 bytes. Each enabled segment is encoded independently; labels do not count. Defaults are illustrative, and actual server usage or billing is not estimated.
Try an example
Compare the capacity left after instructions, retrieved passages and optional conversation history; disable a segment to free space.
How it works
Allocation = sum of enabled segment text tokens + output reserve + user-supplied overhead. Reports per-segment counts, remaining capacity and overflow using the chosen vocabulary.
Frequently asked questions
Allocate a user-defined context window across editable text segments, output reserve and extra overhead.
How do I use Agent Context Budget Planner?
Set your own window and reserves, add labeled segments, enable the content to include, then calculate and export JSON or Markdown.
Does this tool upload my data?
Inputs and files are processed in your browser. Page-view analytics do not contain your tool input.
What should I check before using the result?
Up to 50 segments and 100,000 total UTF-8 bytes. Each enabled segment is encoded independently; labels do not count. Defaults are illustrative, and actual server usage or billing is not estimated.