1C MCP Toolkit

Ten MCP servers for 1C development: platform reference, targeted object dump and load, COM, storage, file search and BSL Language Server

MCP server

High risk

We rate an entry high when the tool writes to external systems, handles money, production databases or secrets, or runs arbitrary commands. The CLI installs it only with your consent.

Why this level

  • load_objects writes changes directly into a work infobase with storage
  • 1c-com reads and writes infobase data over COM under the user's rights
All reasons and checks
Russian stack

adam-rubinstein/1c_mcp

Install

Manual install

python -m venv .venv && pip install -r requirements.txt && python scripts\smoke_test.py

Install dependencies and run an initial environment check on Windows.

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

What it does

The project bundles ten separate MCP servers for specific 1C development tasks: 1c-platform gives platform API reference from an HBK file, 1c-dump and 1c-load do targeted dump and load of configuration objects with integrity checks, 1c-storage works with the configuration storage and history, 1c-com reads and writes infobase data over COM, 1c-files searches already-exported files, 1c-review runs a BSL checklist from YAML patterns, 1c-journal reads the registration log, 1c-debug connects to the HTTP debugger, and 1c-bsl reports BSL Language Server status. For a work infobase there is a mandatory cycle: fetch a storage marker, dump a patch, lock only the needed objects in storage, and only then load with confirmation.

Who it is for. For 1C developers in Cursor and similar environments who want to dump and load configuration objects without manually clicking through the designer at every step.

Good fit when

  • You need to dump a single object without a full configuration export
  • You need a safe storage-capture cycle before loading into a work infobase
  • You need 1C platform API reference right in the editor

Not a fit when

  • You do not have two infobases, a dev sandbox and a work base with storage, which the workflow assumes
  • You need a single simple tool, not ten interconnected servers

Example request

Dump the МойДокумент document form from the dev base and show what changed compared to the repository

Limitations

Built for Windows and the local batch 1cv8 DESIGNER; a targeted dump or load takes designer exclusivity on a file infobase. Secrets live only in a local .env and Cursor settings. Loading into work without a storage marker or without capturing the needed objects is deliberately refused by strict gates.

How to disable. Remove the relevant servers from mcp.json and stop the background processes started by scripts/deploy/start_all.py.

MCP

Transport
stdio, sse
Authentication
not required
Environment variables
Environment variables
ONEC_BIN
required
Path to the 1C platform executable.
ONEC_IB_DEV
required
The dev infobase used as a sandbox.
ONEC_IB_WORK
The work infobase with configuration storage.
ONEC_PASSWORD
required, secret
Password of the 1C user.

Security check

  • load_objects writes changes directly into a work infobase with storage
  • 1c-com reads and writes infobase data over COM under the user's rights

README in short

The README lists ten server packages in a table, describes a two-base dev and work model with a mandatory storage-capture cycle before loading, designer exclusivity during a targeted dump, an adopted UUID check and configuration-root protection, a quick start via smoke_test.py, and a security section stating MCP access to an infobase must not be exposed to the internet without authorization.

FAQ

What happens if objects are not locked in storage?

load returns objectsToCapture instead of a silent success and refuses to load objects that were not explicitly captured.

Can HTTP or SSE be exposed to the internet?

The author explicitly warns that MCP access to the infobase equals access to the configuration and data, so it should not be exposed without authorization.

Editors’ pick

A skills library that gives coding agents a development process: brainstorming, planning, TDD, subagents and code review

PluginMedium riskNo VPN needed292.5KRepository stars
Editors’ pick

Small composable skills for engineering with agents: plan grilling, TDD, bug diagnosis, code review and architecture

SkillLow risk271.4KRepository stars
Editors’ pick

GitHub toolkit for spec-driven development: the specify CLI adds agent commands and skills to a project, from principles to implementation

CLIMedium riskNo VPN needed139.3KRepository stars

Reference MCP servers

Model Context Protocol servers

Official

Official reference MCP servers: Filesystem, Fetch, Git, Memory, Sequential Thinking, Time and Everything

MCP serverMedium risk90.6KRepository stars
Foxx AI1C MCP Toolkit

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