1C MCP Toolkit
A 1C external data processor with MCP and REST API: queries, metadata, event log and code execution without changing the configuration
- 329 ★ on the repository
- Last commit
- GPL-3.0
Broad access
Why this level
- The execute_code tool runs arbitrary code in the 1C base
- Reads live data, including personal data
- The built-in server has no authentication by default
Install
- Works in
- Services
In your terminal, with SkillFoxx CLI
npx skillfoxx add mcp/1c-mcp-toolkitDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- The execute_code tool runs arbitrary code in the 1C base
- Reads live data, including personal data
- The built-in server has no authentication by default
This is third-party code. Review the repository files before installing.
What it does
The tool connects the agent to a 1C infobase through an external data processor, so there is no need to change the configuration or publish the base. In built-in mode the HTTP server runs inside the processor; in proxy mode the processor polls a Python server for commands. The agent runs 1C queries, reads metadata, the event log, access rights and object references, gets BSL syntax help and can execute arbitrary code. Anonymization replaces real values with tokens and restores them in the final answer.
Who it is for. For 1C developers and analysts who need an agent with infobase data access without modifying the configuration.
Good fit when
- You cannot change the configuration or publish the base on a web server
- The agent needs to query data and read the event log
- You need to hide personal data from a cloud model
Not a fit when
- Code execution in the working base is not acceptable
- You need source analysis without a running 1C client
Example request
Use MCP to select the ten latest sales documents with amount and counterpartyMCP
- Transport
- http
- Authentication
- not required
Environment variables
ALLOW_DANGEROUS_WITH_APPROVAL- Allow dangerous operations with approval in proxy mode
ANONYMIZATION_ENABLED- Enable data anonymization in proxy mode
Limitations
The processor must be open in a running 1C client. The built-in server has no authentication by default; a token is enabled on the form. Window screenshots work on Windows only, and on macOS the built-in server needs compiled native components. GPL-3.0 licensed.
How to disable. Stop the server and close the processor in 1C, stop the proxy container and remove the server from the client settings.
FAQ
How do I connect Claude Desktop?
Its config supports only stdio, so the README suggests the mcp-remote bridge via npx or a custom connector on paid plans.
How do I separate test and production bases?
With channel IDs: one proxy, different channel parameters in the URL and processor settings.
Discussion
No questions or reviews yet.
Related
All 1C:Enterprise toolsMCP server inside 1C
1c_mcp
A 1C extension that runs an MCP server inside the infobase: metadata tools out of the box and custom tools in BSL
A 1C extension that runs an MCP server inside the 1C:Enterprise client and lets you write your own tools in BSL
A native component that runs a full MCP server inside 1C: tools, resources and prompts are written in BSL, while a DLL handles the protocol
78 MCP tools for 1C development over local XML exports: metadata, code audit, queries, EPF, configuration diffing, and a verified API reference
1C metadata MCP proxy (adapted fork)
1C_MCP_SERVER_OWN
A Python proxy between an MCP client and a 1C HTTP service: reads metadata, object structure and predefined data, with optional OAuth2
A single-binary Go MCP server: the agent sees 1C metadata, searches code, runs read-only queries and writes accurate BSL
README in short
The README covers two architectures, built-in server and proxy, with a quick start for each including a Docker Hub image. It gives settings for Kiro and Claude Desktop, token authentication, and tables of MCP tools and REST endpoints. Sections cover anonymization, a blocklist of dangerous operations and channel isolation. The skills folder contains skills for the REST API and for composing 1C queries.