AI x 1C Guide
A guide with working scripts on safely connecting an agent to 1C and Bitrix24: OData, webhooks and a catalog of 14 vetted tools
Low risk
We rate an entry low when it mostly gives the agent instructions and reference material.
Why this level
- The examples are read-only by default and do not implement write commands
- The guide explicitly describes write risks via OData and requires separate permissions and negative tests before enabling it
Install
Manual install
python scripts/fresh_odata_example.pyThe 1C:Fresh OData client example, run after cloning the repo and configuring your own connection parameters.
This is third-party code. Review the repository files before installing.
What it does
The repository walks through four real integrations, marking what was verified in practice and what was not: reading and a test write into 1C:UNF via standard OData, reading Bitrix24 tasks through an incoming webhook, creating leads from a website form via a backend webhook, and reconciling requests across 1C-Connect, Jira and Bitrix24. Each integration comes with a working Python script in scripts/, tests, and a guide in guides/. A separate catalog of 14 projects (MCP servers, Agent Skills, hosted services) in catalog/tools.json is tagged with a verification level: Docs, Artifact, CLI smoke or Live smoke, plus license and key risk. The guide sets out a ten-point security baseline and a four-stage rollout order, from source-only access with no database to data writes with human confirmation.
Who it is for. For 1C developers, integrators and managers deciding how to safely give an agent access to 1C or Bitrix24 without risking production data.
Good fit when
- You need a step-by-step breakdown of connecting an agent to 1C:Fresh or UNF OData
- You need to pick an MCP server or Agent Skills for a specific task with an honest risk assessment
- You need a security baseline checklist before giving an agent access to a production base
Not a fit when
- You need a ready MCP server, not a guide with examples and a catalog of links to other projects
- You work with neither 1C nor Bitrix24
Example request
Work out how to safely connect a read-only agent to 1C:Fresh via OData and show an example with my HTTP clientLimitations
The repository itself is not an MCP server or an installable agent tool, but a guide with example scripts and a comparison catalog. Some scenarios (writing to 1C:UNF, part of the Bitrix24 flow) are confirmed in the author's own working code but not publicly reproduced, which the text states honestly. Catalog assessments are based on documentation and the author's local checks, not an independent audit of every project.
How to disable. Simply do not use the example scripts from scripts/; the guide does not install anything into your project automatically.
Security check
- The examples are read-only by default and do not implement write commands
- The guide explicitly describes write risks via OData and requires separate permissions and negative tests before enabling it
README in short
The README gives a quick-start table by scenario, walks through four integrations noting what was verified live, explains that 1C OData is not read-only by default and how to restrict it, provides a tool-selection table and a 14-project catalog table with verification levels and risks, an architecture diagram, four rollout stages, a security baseline and an FAQ.
FAQ
Does the guide guarantee the listed MCP servers are safe?
No, the README states directly it does not declare tools safe or production-ready, only records a verification level.
What is the difference between an MCP server and Agent Skills here?
An MCP server gives an agent tools and protocol access to an external system, while Agent Skills are rules and workflows inside the client itself, often without touching a database at all.
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