# Refract

> Use this tool when you need to efficiently compress tool schemas while preserving signal integrity, reducing token count by up to 98% without relying on large language models (LLMs). It solves problems related to data size and processing speed, providing a fully deterministic output. Ideal for applications requiring fast and reliable schema compression, with zero LLM calls and 100% signal preservation guaranteed.

Canonical page: https://skillsregistry.net/skills/loudilimed-refract  
JSON: https://api.skillsregistry.net/v1/skills/loudilimed-refract

## Description

MCP proxy that compresses tool schemas on the fly. Up to 98% token reduction, 100% signal preserved verified after every compression. Zero LLM calls, fully deterministic.

## Trust

- **Trust score (0–1):** 0.30
- **Verification tier:** scanned
- **Last scanned:** 2026-08-31

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** ai-ml
- **Updated:** 2026-08-31

## Source

- **Source listing:** [Glama](https://glama.ai/mcp/servers/yfzvdnckbh)
- **Repository:** <https://github.com/LoudiliMed/Refract>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "loudilimed-refract"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/loudilimed-refract` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/loudilimed-refract/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt