Skip to main content
Glama
limaduzz11

Protheus Research MCP Server

by limaduzz11
README.md
# Protheus Research MCP Server

MCP (Model Context Protocol) server for technical research in the TOTVS Protheus ecosystem. Enables AI agents to query official documentation, community knowledge bases, and release notes through structured tools.

## Overview

Provides structured access to Protheus technical knowledge — from ADVPL language reference to module-specific documentation, runtime error investigation, and code generation.

## Tools

- **search_protheus_docs** — Search official TOTVS documentation
- **search_community** — Search trusted community sources
- **search_release_notes** — Find release notes and patches
- **compare_sources** — Cross-reference official and community findings
- **deep_research** — Comprehensive multi-source research
- **generate_advpl_example** — Generate ADVPL/TL++ code examples
- **debug_protheus_error** — Investigate runtime errors and stack traces

## Tech Stack

`TypeScript` `Node.js` `MCP SDK` `SQL Server` `REST APIs`

## Quick Start

```bash
# Install dependencies
npm install

# Build
npm run build

# Start MCP server
node dist/index.js
```

## Configuration

Add to your MCP client configuration:

```json
{
  "mcpServers": {
    "protheus-research": {
      "command": "node",
      "args": ["dist/index.js"]
    }
  }
}
```

## License

MIT

TDQS

A4.1/5.0

Scored across 7 tools

Disambiguation4/5

Each tool has a clear primary purpose, and the distinctions between search_protheus_docs, search_community, and search_release_notes are well-defined by source type. There is slight overlap between search_protheus_docs and search_release_notes since release notes are part of official docs, but the descriptions reduce ambiguity.

Naming Consistency5/5

All tool names follow a consistent snake_case verb_noun pattern: search, compare, generate, debug. Even 'deep_research' fits the pattern as a verb-like compound. No mixed conventions or inconsistent casing.

Tool Count5/5

Seven tools is well within the ideal 3-15 range for a focused research and assistance server. Each tool covers a distinct aspect of the Protheus ecosystem, and none feel redundant or unnecessary.

Completeness5/5

The tool surface covers the full research lifecycle: searching official and community sources, checking release notes, comparing sources, running deep aggregated research, generating code examples, and debugging errors. This leaves no obvious dead ends for common Protheus developer tasks.