What is mcp-token-saver CLAUDE.md?

Project instructions for an MCP server that helps compatible coding clients use fewer tokens when working with code and documents.

Use them when developing or testing the mcp-token-saver project, including its token estimates, context optimization, budget checks, and build or test work.

They set a required process for estimating input size, shortening large context, and checking response limits before answering.

How Talap-creator/mcp-token-saver configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to Talap-creator/mcp-token-saver. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/Talap-creator/mcp-token-saver/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/Talap-creator/mcp-token-saver

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/talap-creator/mcp-token-saver/claude-md/github.svg)](https://agentmods.dev/instructions/talap-creator/mcp-token-saver/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/talap-creator/mcp-token-saver/claude-md"><img src="https://agentmods.dev/badge/instructions/talap-creator/mcp-token-saver/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/talap-creator/mcp-token-saver/claude-md"><img src="https://agentmods.dev/badge/instructions/talap-creator/mcp-token-saver/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 592 This file is loaded in full into every session.

When invoked 592 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00592	$0.00592
Opus 5	$0.00296	$0.00296
Sonnet 5	$0.00118	$0.00118
Haiku 4.5	$0.00059	$0.00059

Measured 9d ago against content hash 2c052b81c555, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

mcp-token-saver CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 59 lines

How it starts

The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Project: mcp-token-saver

This project ships an MCP server with three tools — estimate_tokens, optimize_context, check_budget — that any MCP-compatible client (Claude Code included) can call to spend fewer tokens.

Token-saving protocol (MANDATORY)

You have access to the mcp-token-saver MCP server. You MUST follow this protocol on every task:

Before answering any prompt that includes code, file contents, or attached documents, call estimate_tokens with the prompt and the file contents. Pass the active model name.
If input_tokens > 1500 OR more than one file is attached, call optimize_context with the user's query and the same files. Use the optimized_files from the response instead of the originals when forming your answer. Always quote the message field back to the user so they see the savings (e.g. "Context reduced by 65%. ~$0.014 saved.").
Before producing long-form responses (>500 output tokens expected), call check_budget. If warning is low_budget, switch to a concise answer style. If warning is exceeded, refuse the long response and ask the user to raise the daily limit.
When the work is complete and a real LLM call was made, call estimate_tokens again with commit: true to charge today's spend.

Treat these calls as mandatory infrastructure, not optional helpers.

Repo layout

src/index.ts — stdio entrypoint
src/server.ts — tool registration
src/tools/ — estimateTokens.ts, optimizeContext.ts, checkBudget.ts
src/utils/ — tokenizer.ts (Claude/tiktoken router), cost.ts, chunker.ts (Babel AST + blank-line fallback), matcher.ts (keyword scoring), embeddings.ts (local @huggingface/transformers), budget.ts (JSON state with file-lock)
test/ — vitest

Conventions

ESM only ("type": "module"), Node ≥18, TypeScript strict.
No telemetry, no network calls except embedding model download (cached by @huggingface/transformers under ~/.cache).
State file: ~/.mcp-token-saver/state.json (override via MCP_TOKEN_SAVER_STATE).
Daily limit env: MCP_TOKEN_SAVER_DAILY_LIMIT (default 5 USD).

Read the full file on GitHub · 59 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

9d ago First seen · 59 lines · 592 tokens per session scan A 2c052b81c555

Subscribe to this mod's changes

mcp-token-saver CLAUDE.md is an instructions file published in the GitHub repository Talap-creator/mcp-token-saver (6 stars, last pushed 3mo ago), licensed MIT. It adds 592 tokens to every session, about $0.0030 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Claude Code ai anthropic mcp mcp-server promt-engineering

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

mcp-token-saver: Instructions file for Claude Code