1C Templates MCP server
MCP сервер шаблонов 1С
An MCP server for storing and searching your own BSL code templates, with a web editor and syntax highlighting
- 18 ★ on the repository
- Last commit
Extended access
Why this level
- Starts a local HTTP server with MCP and a BSL web console
- The agent can only list, read and search templates, editing happens in the web UI
- Port 8023 is open without auth
Install
- Works in
- Services
Manual install
run.batBuilds the Docker image and starts the container with data/templates mounted.
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Starts a local HTTP server with MCP and a BSL web console
- The agent can only list, read and search templates, editing happens in the web UI
- Port 8023 is open without auth
This is third-party code. Review the repository files before installing.
What it does
A dependency-free pure Python server exposes an MCP endpoint over Streamable HTTP and a web interface with a BSL Console for managing templates. An agent gets three tools: listing templates, fetching a template by id, and fuzzy search over name, description and tags that ignores letter case and supports prefix search for words longer than four characters. Templates are stored as .bsl files plus an index.json with metadata in the data/templates folder, mounted as a Docker volume. About 30 ready-made BSL templates for common tasks ship with it.
Who it is for. For 1C developers who keep a personal library of BSL code templates and want an agent to find and insert them on request.
Good fit when
- You have a set of frequently used BSL code snippets you want an agent to reuse
- You need local template search by name or tags without going online
- A web editor with BSL highlighting for editing templates is convenient
Not a fit when
- You need a shared template base for a team, not a single developer
- You do not use Docker to run local services
Example request
Find my template for exporting a value table to Excel and insert the code into the current moduleMCP
- Transport
- http
- Authentication
- not required
Environment variables
TEMPLATES_DIR- Path to the templates folder inside the container, defaults to data/templates.
Limitations
The project has no license file in the repository. The server is local with no built-in authorization; the web interface and MCP endpoint are reachable by anyone who can hit port 8023. Search is fuzzy and built for short queries, not semantic search.
How to disable. Stop and remove the mcp-templates container with docker rm -f mcp-templates and remove the server from your MCP client config.
FAQ
Do I need Python or pip to run it?
No, the server uses only stdlib Python inside the Docker image, nothing to install locally.
How do I add a template without the web interface?
Drop a .bsl file into data/templates and add an entry to index.json; new files without an entry are picked up automatically on the next list request.
Discussion
No questions or reviews yet.
Related
All 1C:Enterprise toolsAn MCP server with hybrid search over 2262 community BSL code templates and a CRUD web interface built on Monaco Editor
A Dockerized MCP server that keeps BSL Language Server running and gives the agent code navigation, diagnostics and refactoring for 1C instead of grep search
A Dockerized MCP server for searching 1C dumps: structural index, call graph and semantic code search
HTTP 1C MCP gateway with a dashboard
onec-mcp-universal
A Docker gateway with 40-plus MCP tools for 1C, one address, a settings dashboard and a set of 1C skills for Claude Code
A single-binary Go MCP server: the agent sees 1C metadata, searches code, runs read-only queries and writes accurate BSL
MCP bridge to BSL Language Server: navigation, diagnostics and refactoring of 1C and OneScript code for agents
README in short
The README describes launching via run.bat or manually via docker build and docker run with data/templates mounted, gives a connection config for Cursor and Claude Code, a table of three MCP tools, and a usage example in Cursor. It separately explains template storage and the project structure: plain Python with no dependencies.