1c-configuration-indexer

A local CLI indexer for 1C XML and BSL exports that chunks a configuration and uploads it to an external backend for semantic search

CLI

Medium risk

We rate an entry medium when the tool runs code, makes network calls or reads project files. Check what exactly it does before installing.

Why this level

  • Sends configuration code and metadata fragments to an external webhook and Supabase outside the user's own infrastructure
  • Requires a token to access the external upload service
All reasons and checks
Russian stack

tarasov46/1c-configuration-indexer

Install

Manual install

python -m venv .venv && .\.venv\Scripts\python.exe -m pip install -e .

Installing the indexer from source on Windows.

This is third-party code. Review the repository files before installing.

What it does

The tool takes a local folder with a 1C configuration export (src and optionally exchanges with extensions), builds a compact index of navigation entities, RAG chunks and the relations between them, and, following a job from configuration-mcp, uploads the packaged chunks over HTTP to an external webhook that feeds them into Supabase. Full BSL and query text is not the source of truth: Supabase gets the path, line numbers, hashes, lengths and short previews, while the exact code stays in Git. It includes commands to retry failed upload parts, rebuild a package with smaller chunks, and run a local embedding worker as a manual fallback to the scheduled n8n or NeuroCore worker.

Who it is for. For teams already using a configuration-mcp and Supabase pairing for semantic search over 1C configurations, who need a local indexing and upload step.

Good fit when

  • You already run a configuration-mcp and Supabase setup and need a local indexer for it
  • You need to index a configuration or individual extensions without passing large files into the model's context

Not a fit when

  • You do not have your own configuration-mcp, Supabase project and the author's webhook infrastructure running: the tool targets that specific setup rather than working standalone
  • You need local semantic search without sending data out

Example request

Index the extension in the exchanges folder and upload the package per the configuration-mcp job

Limitations

The tool sends configuration chunks over HTTP to a specific webhook on the author's infrastructure (d4vibe.core.neuro42.ru) and into Supabase; without access to that infrastructure or your own equivalent, the upload-package and run-job upload commands are not useful. The local embedding worker optionally uses the OpenAI API. The README explicitly asks not to commit 1C exports, generated packages, .cf, .cfe, .dt files or tokens.

How to disable. Delete the project's virtual environment and stop running configuration-indexer.exe; the tool leaves no local background processes.

Security check

  • Sends configuration code and metadata fragments to an external webhook and Supabase outside the user's own infrastructure
  • Requires a token to access the external upload service

README in short

The English README describes an autonomous flow from a job file to an uploaded package, a project layout with src and exchanges, a compact v2 package format with entities, search chunks and relations, separating upload from embedding generation via a scheduled worker, the upload-package, retry-failed-package and rechunk-package commands with a specific webhook address on the author's infrastructure, and a safety rule against committing exports and tokens.

FAQ

Is full BSL code stored in Supabase?

No, Git remains the source of truth for code; Supabase gets the path, hashes, lengths and short previews.

Can indexing be tested without uploading?

Yes, the --no-upload flag on run-job builds the package locally without sending it.

Editors’ pick

166 skills for scientific work: bioinformatics, cheminformatics, clinical data, geospatial analysis and 100+ databases

SkillMedium risk47KRepository stars
Official

Google's open-source MCP server for databases: ready tools for Postgres, MySQL, BigQuery, Spanner and more, plus custom tools in tools.yaml

MCP serverHigh risk16.5KRepository stars
Editors’ pick

Official Hugging Face skills: Hub operations via the hf CLI, datasets, model training, Spaces, evals and deployment

SkillHigh risk11.1KRepository stars
Official

A visualization language for agents and an MCP server: neat charts from a simple semantic spec

MCP serverMedium riskNo VPN needed4.3KRepository stars
Foxx AI1c-configuration-indexer

I am Foxx AI and I have already vetted this tool. Ask about install, setup or anything else, and I will keep it simple.