<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/Yk46q4wPy" -->

---
title: T-Zero: An open-source tool that cuts LLM prompt token...
description: T-Zero is a free, open-source MCP server that cuts LLM prompt token costs by stripping routine function bodies from code context. It uses Python AST parsing to...
canonical: https://daily.dev/posts/t-zero-an-open-source-tool-that-cuts-llm-prompt-token-costs-by-up-to-95-25k-downloads--yk46q4wpy
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: T-Zero: An open-source tool that cuts LLM prompt token costs by up to 95% (25k+ downloads) | daily.dev
og:description: T-Zero is a free, open-source MCP server that cuts LLM prompt token costs by stripping routine function bodies from code context. It uses Python AST parsing to...
og:url: https://daily.dev/posts/t-zero-an-open-source-tool-that-cuts-llm-prompt-token-costs-by-up-to-95-25k-downloads--yk46q4wpy
og:image: https://api.daily.dev/og/posts/Yk46q4wPy.png
og:image:alt: T-Zero: An open-source tool that cuts LLM prompt token costs by up to 95% (25k+ downloads)
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# T-Zero: An open-source tool that cuts LLM prompt token costs by up to 95% (25k+ downloads)

**[Dev Tools](https://daily.dev/sources/devtools)** · [@toprak4](https://daily.dev/toprak4) · 2 min read · 15 upvotes · 2 comments

## Summary

T-Zero is a free, open-source MCP server that cuts LLM prompt token costs by stripping routine function bodies from code context. It uses Python AST parsing to remove implementation details while preserving type signatures, class hierarchies and symbol links, claiming 85% to 95% token reduction. It runs 100% offline with zero telemetry and includes a built-in AST secret scanner to stop API keys leaking into prompts. It ships 20 MCP tools, passes 105 automated unit tests, and has surpassed 25,000 downloads across PyPI and npm.

## Content

Hey everyone! 👋

When working with LLM agents (Cursor, Windsurf, Claude Desktop, VS Code), context windows get clogged by routine implementation bodies—exploding API token bills and slowing down inference.

To solve this, I built **T-Zero** — a 100% free and open-source context architect and canonical Model Context Protocol (MCP) server.

### 🚀 What T-Zero does:

- **AST Interface Skeletonization:** Uses pure Python AST parsing to strip routine function bodies while preserving complete type signatures, class hierarchies, and symbol links (**85%–95% token reduction**).
- **100% Local & Zero-Leak:** Zero telemetry and 100% offline execution. Includes a built-in AST secret scanner so proprietary API keys never leak into LLM prompts.
- **Production-Grade:** 20 integrated MCP tools, 105 automated unit tests passing, certified TDQS 5.0/5.0 Grade A.
- **Traction:** Just surpassed **25,000+ downloads** across PyPI (`tzero-mcp`) and npm.

### 🛠️ Quick Try:

```
# Install via pip:
pip install --upgrade tzero-mcp

# Or 1-click IDE setup:
npx tzero-mcp --install
```

### 🔗 Explore & Benchmark:

- ⭐ **GitHub:** https://github.com/toprakahmetaydogmus/TZeroAlgorithm
- 🌐 **Live Demo & Benchmarks:** https://tzero-indol.vercel.app

I'd love your feedback, code reviews, and benchmark tests on your own projects! If T-Zero helps optimize your context budget, **dropping a star on GitHub would be amazing support!** ⭐🙏

## Community discussion

Top comments from developers on daily.dev.

**@bestseo** · 1 upvotes

> Sounds great, and thanks for making it free and open source, @toprak4 .
> How can we ensure that T-Zero retains all the context and memory it needs between sessions?

## Similar posts on daily.dev

- [GitHub - knowsuchagency/mcp2cli: Turn any MCP server or OpenAPI spec into a CLI — at runtime, with zero codegen](https://daily.dev/posts/github---knowsuchagency-mcp2cli-turn-any-mcp-server-or-openapi-spec-into-a-cli-at-runtime-with-z-olbhdwpeq) · Hacker News · 1 upvotes · 0 comments
- [One MCP server used 18,000 tokens before doing anything. Here’s the workaround.](https://daily.dev/posts/one-mcp-server-used-18-000-tokens-before-doing-anything-here-s-the-workaround--z0ugekuke) · The New Stack · 1 upvotes · 0 comments
- [Medium](https://daily.dev/posts/medium-4ldzlfdn3) · Medium · 0 upvotes · 0 comments
- [MCP tool discovery for autonomous LLM agents](https://daily.dev/posts/mcp-tool-discovery-for-autonomous-llm-agents-a9b2gncqd) · portkey · 2 upvotes · 0 comments
- [Tokensparsamkeit for coding assistants](https://daily.dev/posts/tokensparsamkeit-for-coding-assistants-f7i9ttrv5) · ITNEXT · 0 upvotes · 0 comments

---

Tags: [#python](https://daily.dev/tags/python), [#open-source](https://daily.dev/tags/open-source), [#ai-agents](https://daily.dev/tags/ai-agents), [#mcp](https://daily.dev/tags/mcp), [#context-engineering](https://daily.dev/tags/context-engineering)

[View this post on daily.dev](https://daily.dev/posts/t-zero-an-open-source-tool-that-cuts-llm-prompt-token-costs-by-up-to-95-25k-downloads--yk46q4wpy)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"DiscussionForumPosting","mainEntityOfPage":"https://daily.dev/posts/t-zero-an-open-source-tool-that-cuts-llm-prompt-token-costs-by-up-to-95-25k-downloads--yk46q4wpy","headline":"T-Zero: An open-source tool that cuts LLM prompt token costs by up to 95% (25k+ downloads)","text":"T-Zero is a free, open-source MCP server that cuts LLM prompt token costs by stripping routine function bodies from code context. It uses Python AST parsing to remove implementation details while preserving type signatures, class hierarchies and symbol links, claiming 85% to 95% token reduction. It runs 100% offline with zero telemetry and includes a built-in AST secret scanner to stop API keys leaking into prompts. It ships 20 MCP tools, passes 105 automated unit tests, and has surpassed 25,000 downloads across PyPI and npm.","url":"https://daily.dev/posts/t-zero-an-open-source-tool-that-cuts-llm-prompt-token-costs-by-up-to-95-25k-downloads--yk46q4wpy","datePublished":"2026-11-19T14:11:10.881Z","dateModified":"2026-11-19T14:11:30.901Z","author":{"@type":"Person","name":"Toprak","url":"https://daily.dev/toprak4","image":"https://lh3.googleusercontent.com/a/ACg8ocL086UPBGns3t9h4hfJUeR5w9G5mWsLy8Ub_GfBzFO_UndQx9oL=s96-c","interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"EndorseAction"},"userInteractionCount":60}},"interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":15},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":3}],"comment":[{"@type":"Comment","text":"Sounds great, and thanks for making it free and open source, @toprak4 .\nHow can we ensure that T-Zero retains all the context and memory it needs between sessions?","datePublished":"2026-11-20T06:39:53.528Z","url":"https://daily.dev/posts/Yk46q4wPy#c-yp1nu8EVx","author":{"@type":"Person","name":"Albert Adaikkalam","url":"https://daily.dev/bestseo","image":"https://avatars.githubusercontent.com/u/608213?v=4"},"interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1}},{"@type":"Comment","text":"Guys. ı will be honest ım 15 y.o developer by myself. there is no one helping me to improve this project. ı added a sponsor thing to the repo ı wanna keep this project for free but ı need a motivation can you guys help ?","datePublished":"2026-11-20T15:35:57.760Z","url":"https://daily.dev/posts/Yk46q4wPy#c-8GhKTEOBR","author":{"@type":"Person","name":"Toprak","url":"https://daily.dev/toprak4","image":"https://lh3.googleusercontent.com/a/ACg8ocL086UPBGns3t9h4hfJUeR5w9G5mWsLy8Ub_GfBzFO_UndQx9oL=s96-c"}}],"isPartOf":{"@type":"WebPage","url":"https://daily.dev/squads/devtools","name":"Dev Tools"}}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Dev Tools","item":"https://daily.dev/squads/devtools"},{"@type":"ListItem","position":3,"name":"T-Zero: An open-source tool that cuts LLM prompt token costs by up to 95% (25k+ downloads)"}]}
```

