Docker deployment for 1C MCP servers
MCP-серверы для 1С:Enterprise
Ready Docker Compose files and management scripts for nine 1C:Enterprise MCP servers: help, metadata, graph search, syntax checking
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
- The servers get access to 1C configuration metadata, code and reference data
- Requires license keys and passwords in .env, kept out of git
Install
Manual install
cd mcp-deployment && cp .env.example .env && ./scripts/setup.sh && ./scripts/start.shRequires license keys for the 1C MCP servers and an OpenRouter or OpenAI API key for embeddings.
This is third-party code. Review the repository files before installing.
What it does
The repository does not contain the MCP servers themselves; it packages ready configuration to run them in Docker: compose files for the base set and separately for Neo4j-based graph search, FormsServer and cloud embeddings, plus setup, start, stop, restart, status, logs and reindex scripts. The servers run from comol/* images with no build step, but the 1C:Enterprise MCP servers themselves are licensed and purchased separately at vibecoding1c.ru. After a 1C configuration update, reindex.sh rebuilds the indexes of the chosen servers.
Who it is for. For 1C developers who already bought or plan to buy the licensed vibecoding1c.ru MCP servers and want to deploy them through Docker Compose with ready management scripts.
Good fit when
- You have the licensed vibecoding1c.ru 1C MCP servers and need a ready Docker deployment
- You need separate scripts for start, status and reindexing after a configuration change
- You need Neo4j-based graph search over metadata alongside the base servers
Not a fit when
- You have no plan to buy the licensed vibecoding1c.ru servers: the config is useless without them
- You need a production environment: the author explicitly warns the config is not tested for production
Example request
Check this BSL module's syntax with the 1c-syntax server and find a matching code templateLimitations
The 1C:Enterprise MCP servers themselves are not included and are bought separately under a license at vibecoding1c.ru; without the license keys the config will not start. The README explicitly says it is not for production systems, has not been fully tested, and is used at your own risk. Some servers share port 8000 and cannot run at the same time.
How to disable. Stop the containers with ./scripts/stop.sh and remove the server entries from your MCP client config.
MCP
- Transport
- http
- Authentication
- API key
| Environment variables | |
|---|---|
| CODE_PATH required | Path to the 1C configuration exported to files. |
| METADATA_HOST_PATH required | Path to the configuration report used for indexing. |
Security check
- The servers get access to 1C configuration metadata, code and reference data
- Requires license keys and passwords in .env, kept out of git
README in short
The README warns upfront that it needs the licensed 1C MCP servers, bought separately at vibecoding1c.ru. It then lists nine servers with ports and purposes: platform help, metadata graph search, code search, standard subsystems help, templates, syntax checking, 1C:Napparnik-based checking, forms context, and cloud embeddings. It describes the reindexing process after a configuration or extension update.
FAQ
Do I need to build the images myself?
No, the servers run from ready comol/* images with no build step, but you need license keys.
Can I connect only some of the servers?
Yes, the README explicitly advises connecting only the servers you need and removing the rest from the MCP client config.
Related
An MCP server built into the Netdata agent: metrics, logs, alerts and live process, service and container data for an AI assistant
GitHub's official MCP server: code, issues, pull requests, Actions and security alerts straight from the agent
Agent Skills for Google products
Agent Skills for Google products and technologies
Official Google skill collection for working with Google Cloud, BigQuery, GKE, ads and analytics from an agent
AWS's official MCP server suite: docs, IaC, containers, serverless, databases, cost and monitoring