Skip to main content

Harness Context-Budget Planner

See how many context-window tokens your CLAUDE.md/AGENTS.md, skills, and tool schemas consume before a user types anything — split into always-on vs on-demand, with trim suggestions.

100% local — nothing you paste here is ever uploaded
Try a Sample Harness
Instructions File (CLAUDE.md / AGENTS.md)
Paste your file content0 lines · ~0 tokens
Skills
Tool / Function Schemas (optional)
Paste tool definitions (JSON)0 tokens
Context Window Reference
Characters per token (estimate)
ModelContext Window (tokens)
Defaults are the published limits as of November 2026 and change often; some providers charge more or apply different rate limits for very long prompts. Edit any row to match your model's current documentation.
Always-On Budget vs Context Window
Token Breakdown
0%
of 's context window is consumed by always-on tokens (instructions file + skill descriptions + tool schemas) before the first user message.
Trim Suggestions

Frequently Asked Questions

Is my CLAUDE.md or AGENTS.md content uploaded anywhere?

No. Everything you paste is counted and analyzed in your browser with JavaScript. Nothing is sent to a server, logged or stored, and closing the tab discards it.

How accurate are the token counts?

They are estimates from a characters-per-token ratio you can change. The default of 2.5 is the conservative figure Anthropic documents for current Claude models (about 2.5 million characters per 1 million tokens); 4 is the usual English-prose rule of thumb for GPT and Gemini tokenizers. Code, JSON and non-English text usually need more tokens per character. Use the numbers to compare parts of a harness and spot bloat. For exact counts use the provider's counter: Anthropic's count_tokens endpoint, OpenAI's tiktoken, or Gemini's countTokens.

What is the difference between always-on and on-demand tokens?

Always-on tokens are in the context window on every turn: your instructions file (CLAUDE.md or AGENTS.md), tool and function schemas, and the short name and description of each skill. On-demand tokens, mainly a skill's full body, load only when the skill is used. Always-on size is what you pay for before the user types anything and what competes with the task for the model's attention.

How large should CLAUDE.md or AGENTS.md be?

As small as it can be while still carrying what the agent cannot discover from the repository itself: build and test commands, conventions that differ from the defaults, and hard constraints. Move rarely needed procedures into skills, which load on demand, or into docs the agent reads when relevant. The planner flags long files and long skill descriptions with trim suggestions so you can see which sections cost the most.

What are the context window sizes of current models?

As published in November 2026: Claude Opus 5.5, Sonnet 5.5 and Haiku 5.5 each have a 1M-token context window; GPT-5.5 has 1,050,000 tokens (prompts over 272K input tokens are priced higher); Gemini 3.1 Pro has a 1M-token window. These change often, so the reference table is editable and the provider documentation is authoritative.