kd-rules-mcp

An MCP server for 1C exchange rules: an agent reads, builds, edits and validates Data Conversion 2 and 3 rules

  • 30 ★ on the repository
  • Last commit
  • MIT
MCP server

Extended access

Why this level

  • The server writes exchange rules and extensions into the 1C project folders listed in `projects.yaml`.
  • The container gets access to the project and exchange folders on the local machine.
All reasons and checks
Russian stack

egordoronchenko/kd-rules-mcp

Install

Manual install

docker compose up -d

Before launch pick a release tag, download `docker-compose.yml` and the `projects.example.yaml` samples, run the image `setup` and fill in `projects.yaml`.

Access and checks

How we review

Extended accessThe tool runs code or makes network calls and only reads external data.

  • The server writes exchange rules and extensions into the 1C project folders listed in `projects.yaml`.
  • The container gets access to the project and exchange folders on the local machine.

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

What it does

The server reads 1C configuration XML exports, with no database data needed, and works with exchange rules in two generations of Data Conversion. For Data Conversion 2 it reads and writes `ПравилаОбмена` in the 2.01 format, proposes mapping candidates and edits conversion, property and registration rules. For Data Conversion 3 on the universal EnterpriseData format it parses the exchange manager module, checks it against the format schema and the configuration, extends the standard exchange with an extension and carries registration rules over from version 2. The server is deterministic: the agent decides what converts into what, and the result shows in the report and the `rules_diff` comparison. The `rules_validate` check finds, for example, value types with no conversion rule. Four skills sit in the repository next to the server: installation, Data Conversion 2 exchange pitfalls, Data Conversion 2 rule building and Data Conversion 3 rules.

Who it is for. 1C developers who set up exchange between configurations and want to manage rules through an agent instead of the Data Conversion tool

Good fit when

  • You want to add a catalog to an exchange and build mapping rules from the metadata of two configurations.
  • Counterparties get duplicated during exchange and you are tracing the cause in the conversion rules.
  • A standard Data Conversion 3 exchange is extended and its exchange manager module must be checked against the format schema.
  • Registration rules from Data Conversion 2 are being moved to the universal format.

Not a fit when

  • You edit rules on production databases: the server works from exports, and live exchange is checked only between sandboxes.
  • There is no Docker on the machine and you do not want to run the server from a clone through `uv`.
  • The exchange is built without Data Conversion and EnterpriseData.

Example request

Add the Nomenclature catalog to the exchange with Accounting, show the rules diff, then run validation

MCP

Transport
http
Authentication
not required

Limitations

The ready image needs Docker with Compose, and building from a clone needs Python 3.12 and `uv`. The optional final live-exchange checks need Windows with the 1C 8.3 platform. Before launch you fill in `projects.yaml` and `projects.local.yaml` with absolute paths on your machine. The server only parses exports and rules while the agent makes mapping decisions, so the result is worth reviewing in the diff comparison. The author states that the checks do not touch production databases.

How to disable. Stop the container with `docker compose down` and remove the server from `.mcp.json` or the Cursor MCP settings.

FAQ

How do I install kd-rules-mcp in Claude Code?

Bring the container up from the release image with `docker compose up -d`, then run `claude` in the folder and approve the servers from `.mcp.json`. Check the connection with `claude mcp list`.

Do I need a 1C database and the Data Conversion tool?

No. The server reads XML configuration exports, and the Data Conversion product is not required.

How do Data Conversion 3 and 2 differ for the server?

For Data Conversion 2 the server works with `ПравилаОбмена` in the 2.01 format; for Data Conversion 3 it parses the exchange manager module on the universal EnterpriseData format.

What does rule validation show?

The `rules_validate` command prints the number of errors and warnings with a code and location, for example a missing PKO for catalog types. After a fix, `rules_diff` shows what changed.

Which skills come in the box?

Four: `kd-install` for installation, `kd2-exchange-pitfalls`, `kd2-rules-build` and `kd3-rules`. They are packed into release archives for Claude Code, Cursor, OpenCode and Codex.

Discussion

No questions or reviews yet.

Russian stack

An MCP server that parses 1C exchange rules in Data Conversion 2.0 format and answers questions about them in chat, with no LLM involved in the parsing itself

MCP serverExtended access5Repository stars
Russian stack

Rules, subagents, skills and MCP integrations for 1C:Enterprise 8 development with Cursor, Claude Code, Codex and other agents

PluginBroad access539Repository stars
Russian stack

A Docker server with a read-only core and pluggable modules: metadata, code, forms and relations of several 1C configurations through one MCP contract

MCP serverExtended access12Repository stars
Russian stack

An MCP server and portable admin tool for analyzing an exported 1C XML configuration: object listing, metadata structure, code search

MCP serverExtended access2Repository stars

1C Development Standards MCP

MCP-сервер стандартов разработки 1С:Предприятие 8

Russian stack

An MCP server exposing 1C:Enterprise 8 development standards for agent-driven search and code review

MCP serverExtended access3Repository stars
Russian stack

MCP server for 1C:Enterprise databases over standard OData: catalogs, documents, registers and accounting with no configuration extension

MCP serverBroad access13Repository stars

README in short

The README shows a rule validation example, lists capabilities for Data Conversion 2 and 3 and gives two install routes: a ready Docker image and a source build. Skills and hooking into a 1C project folder are described separately.

README badge

Report a problem