1creader: read a 1C file infobase
1creader
A read-only MCP and CLI that read a 1C file infobase directly from 1Cv8.1CD and build a configuration map for an AI agent and Obsidian
Low risk
We rate an entry low when it mostly gives the agent instructions and reference material.
Why this level
- The tool is explicitly read-only, write access is not supported
- The MCP returns catalog and register data, which can include personal data stored in the database
Install
Manual install
git clone https://github.com/vengeoff/1creader.git
cd 1creader
python3 -m venv .venv && .venv/bin/pip install -e .Install from source.
This is third-party code. Review the repository files before installing.
What it does
The tool reads the 1Cv8.1CD file of a file-based 1C infobase without the platform and without an XML export, parsing metadata, BSL module text, forms, and catalog and register data. From that it builds a compact MAP.md for an agent, an Obsidian vault with relation and subsystem maps under vault/, full schemas under developer/, and RAG chunks in ai/retrieval.json. The same MCP server connects to Cursor, Claude Code and Codex, with tools for the index, object schemas, modules, catalog data and Obsidian notes. A parallel onecreader CLI offers info, ls, obj, module, search, form and catalog-data commands.
Who it is for. For anyone handed an unfamiliar or legacy 1C file infobase with no documentation who wants to understand its structure quickly, without the platform and without an XML export.
Good fit when
- You need to understand an unfamiliar or legacy 1C configuration from a 1Cv8.1CD file
- You need a configuration map for an agent in Cursor, Claude Code or Codex without an EDT or XML export
- You want a navigable Obsidian base with subsystem and object relation maps
Not a fit when
- You need to write changes back to the database: the tool only reads
- The file infobase is in a 1CD format newer than 8.3.8, which the project does not yet support
- The database is held open by the Configurator and the file is locked for reading
Example request
Show the module of the Nomenclature catalog and explain how it relates to order documentsLimitations
Only the 1CD 8.3.8 format and platform 8.3.8+ are supported; newer file formats are not claimed. The tool only reads the database, write access is not supported in any mode. The file infobase must be accessible as a file on disk, and the Configurator can hold a lock that blocks reading. Python 3.10 or newer is required.
How to disable. Remove the 1creader entry from .mcp.json, .cursor/mcp.json or .codex/config.toml, and delete the .onecreader artifact folders if you no longer need them.
MCP
- Transport
- stdio
- Authentication
- not required
Security check
- The tool is explicitly read-only, write access is not supported
- The MCP returns catalog and register data, which can include personal data stored in the database
README in short
The README explains why reading the file infobase directly is useful, gives a quick start with the onecreader init command, describes the artifact layout (MAP.md, vault, developer, ai) and the Obsidian vault structure, shows MCP connection snippets for Cursor, Claude Code and Codex, a table of seven MCP tools, and a list of CLI commands. It ends with what gets read (catalogs, documents, registers, modules, database data, DBSCHEMA) and honest limitations on format support and the read-only mode.
FAQ
Does it need the 1C platform installed?
No, the tool reads the 1Cv8.1CD file directly and does not require the platform to be installed.
Can catalog data be read, not just the structure?
Yes, the config_catalog_data tool and the catalog-data command read data from the database tables, not just the object schema.
Related
A skills library that gives coding agents a development process: brainstorming, planning, TDD, subagents and code review
Skills for real engineers by Matt Pocock
Skills For Real Engineers
Small composable skills for engineering with agents: plan grilling, TDD, bug diagnosis, code review and architecture
GitHub toolkit for spec-driven development: the specify CLI adds agent commands and skills to a project, from principles to implementation
Reference MCP servers
Model Context Protocol servers
Official reference MCP servers: Filesystem, Fetch, Git, Memory, Sequential Thinking, Time and Everything