Arman-Kudaibergenov

Arman-Kudaibergenov / bsl-atlas

Public

MCP-сервер для 1С/BSL: векторный поиск, структурный индекс, граф вызовов

18
1
100% credibility
Found Mar 01, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

BSL Atlas is a search tool for 1C:Enterprise configurations that enables fast structural queries for functions, call graphs, and metadata alongside semantic code searches via AI assistants.

How It Works

1
🔍 Discover the helper

You learn about BSL Atlas, a smart tool that helps you quickly search and understand your 1C projects without digging through files.

2
📤 Save your project files

In your 1C program, export your configuration to a folder on your computer, creating a bunch of code and data files.

3
🚀 Set up the search map

Download the easy setup package, point it to your project folder, and let it create a fast map of all functions, data, and connections.

4
▶️ Start the tool

Launch it once, and it builds everything in the background while you grab a coffee.

5
🔗 Link to your AI buddy

Connect it simply to your AI assistant like Claude, so now your chat can use the smart map.

6
💡 Ask and explore

Chat with your AI: 'Show me the posting function' or 'What calls this procedure?' and get instant code, graphs, and details.

🎉 Navigate like a pro

Your huge 1C project feels small and simple now, saving you hours of searching every time you work on it.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is bsl-atlas?

BSL Atlas is a Python MCP server that indexes 1C Enterprise config dumps for vector search, structural queries, and call graphs over BSL code and metadata. Export your configuration files from 1C Configurator, run it via Docker Compose, and AI tools like Claude query functions, modules, or semantics like "error logging code" instantly—no raw file parsing needed. It delivers SQLite speed for exact matches and ChromaDB for natural language hits.

Why is it gaining traction?

Dual indexing shines: SQLite rebuilds in seconds for structural tools like function lookups and metadata search, while vector layer handles fuzzy queries with hybrid cloud/local embeddings (OpenRouter init, Ollama queries). MCP endpoints plug straight into Claude Desktop, VSCode, or IntelliJ for mcp github copilot vscode flows, plus reindex API keeps it fresh post-changes. Low-cost local mode hooks devs tired of cloud bills.

Who should use this?

1C/BSL maintainers navigating sprawling enterprise configs with Claude AI assistants. Teams using mcp github copilot intellij for call graph analysis or mcp github issues triage in Russian-heavy codebases. Config admins querying metadata like "register dimensions" without Configurator hunts.

Verdict

Strong early bet for atlas bsl users—Docker-ready with clear MCP tools and health checks, but 18 stars and 1.0% credibility score mean it's pre-production; test on a real dump before relying. Grab it if MCP python servers fit your 1C workflow.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.