1C AI DevStart

A Claude Code starter template for 1C: 16 skills, 6 role subagents and a protocol for checking against a live base via MCP

Workflow

Medium risk

We rate an entry medium when the tool runs code, makes network calls or reads project files. Check what exactly it does before installing.

Why this level

  • Batch scripts can export and import the configuration through the real Configurator
  • Agent permissions are configured by the user and cautious by default, but can be widened to direct writes to source files
All reasons and checks
Russian stack

mr-ske1r/1c-ai-devstart

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add workflows/1c-ai-devstart

Detects the agents on your machine, checks the risk and pins the version.

Other ways to install

Assembled automatically, review before installing.

Run in a terminal in the project folder

npx skills add mr-ske1r/1c-ai-devstart --skill 1c-batch 1c-feature-dev 1c-integrations 1c-odata brainstorm config-modification enterprisedata-exch errors extract-to-extension form-work mcp-1c obj-lifecycle query-writing skd-design ssl-ext-commands use-bsp -a claude-code -y

The skills tool installs the current version from the repository. Add the -g flag to use the skill in every project.

Add the rules to the project

curl -fsSL https://raw.githubusercontent.com/mr-ske1r/1c-ai-devstart/a31d40a7c9b62baeeb42c2204116680664998da6/.claude/rules/1c-rules.md >> AGENTS.md

If AGENTS.md already exists, the rules are appended to the end. Commands for macOS and Linux, on Windows run them in Git Bash.

Add the rules to the project

curl -fsSL https://raw.githubusercontent.com/mr-ske1r/1c-ai-devstart/a31d40a7c9b62baeeb42c2204116680664998da6/.claude/rules/common-rules.md >> AGENTS.md

If AGENTS.md already exists, the rules are appended to the end. Commands for macOS and Linux, on Windows run them in Git Bash.

Add the rules to the project

curl -fsSL https://raw.githubusercontent.com/mr-ske1r/1c-ai-devstart/a31d40a7c9b62baeeb42c2204116680664998da6/.claude/rules/mcp.md >> AGENTS.md

If AGENTS.md already exists, the rules are appended to the end. Commands for macOS and Linux, on Windows run them in Git Bash.

Run in a terminal in the project folder

npx skills add mr-ske1r/1c-ai-devstart --skill 1c-batch 1c-feature-dev 1c-integrations 1c-odata brainstorm config-modification enterprisedata-exch errors extract-to-extension form-work mcp-1c obj-lifecycle query-writing skd-design ssl-ext-commands use-bsp -a codex -y

The skills tool installs the current version from the repository. Add the -g flag to use the skill in every project.

Add the rules to the project

curl -fsSL https://raw.githubusercontent.com/mr-ske1r/1c-ai-devstart/a31d40a7c9b62baeeb42c2204116680664998da6/.claude/rules/1c-rules.md >> AGENTS.md

If AGENTS.md already exists, the rules are appended to the end. Commands for macOS and Linux, on Windows run them in Git Bash.

Add the rules to the project

curl -fsSL https://raw.githubusercontent.com/mr-ske1r/1c-ai-devstart/a31d40a7c9b62baeeb42c2204116680664998da6/.claude/rules/common-rules.md >> AGENTS.md

If AGENTS.md already exists, the rules are appended to the end. Commands for macOS and Linux, on Windows run them in Git Bash.

Add the rules to the project

curl -fsSL https://raw.githubusercontent.com/mr-ske1r/1c-ai-devstart/a31d40a7c9b62baeeb42c2204116680664998da6/.claude/rules/mcp.md >> AGENTS.md

If AGENTS.md already exists, the rules are appended to the end. Commands for macOS and Linux, on Windows run them in Git Bash.

Run in a terminal in the project folder

npx skills add mr-ske1r/1c-ai-devstart --skill 1c-batch 1c-feature-dev 1c-integrations 1c-odata brainstorm config-modification enterprisedata-exch errors extract-to-extension form-work mcp-1c obj-lifecycle query-writing skd-design ssl-ext-commands use-bsp -a opencode -y

The skills tool installs the current version from the repository. Add the -g flag to use the skill in every project.

Add the rules to the project

curl -fsSL https://raw.githubusercontent.com/mr-ske1r/1c-ai-devstart/a31d40a7c9b62baeeb42c2204116680664998da6/.claude/rules/1c-rules.md >> AGENTS.md

If AGENTS.md already exists, the rules are appended to the end. Commands for macOS and Linux, on Windows run them in Git Bash.

Add the rules to the project

curl -fsSL https://raw.githubusercontent.com/mr-ske1r/1c-ai-devstart/a31d40a7c9b62baeeb42c2204116680664998da6/.claude/rules/common-rules.md >> AGENTS.md

If AGENTS.md already exists, the rules are appended to the end. Commands for macOS and Linux, on Windows run them in Git Bash.

Add the rules to the project

curl -fsSL https://raw.githubusercontent.com/mr-ske1r/1c-ai-devstart/a31d40a7c9b62baeeb42c2204116680664998da6/.claude/rules/mcp.md >> AGENTS.md

If AGENTS.md already exists, the rules are appended to the end. Commands for macOS and Linux, on Windows run them in Git Bash.

You will need: Node.js

Checked against the repository on Sep 26, 2026, commit a31d40a.

Text for your agent

Copy the mr-ske1r/1c-ai-devstart repository files (except README.md and install.md) into the 1C project, open Claude Code there and run /setup-project, then restart the session so the MCP servers from .mcp.json come up.

Other ways from the author
Text for your agent

Установи шаблон по инструкции https://raw.githubusercontent.com/mr-ske1r/1c-ai-devstart/master/install.md

Handed to the agent directly in Claude Code, opened in the future project's folder.

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

What it does

The template is copied into a 1C project and set up with a /setup-project command that connects to the database and MCP, exports source files, fills in the CLAUDE.md project passport, and configures agent permissions (what it can change on its own, what needs an explicit request, what is read-only). It provides 16 skills covering topics like queries, SKD reports, managed forms, object lifecycle, EnterpriseData exchanges and batch Configurator operations, plus 6 narrow role subagents: architect, code explorer, developer, reviewer, simplifier, documenter. The key principle: coding rules live in one 1c-rules.md file, and the agent pulls metadata and data from a live base through the third-party 1c_mcp MCP server by Vladimir Kharin, rather than guessing from memory. A /1c-feature-dev command drives the full task cycle, while /new-brief plus /new-plan cover simpler tasks.

Who it is for. For Windows-based 1C developers who want a systematic Claude Code framework instead of a pile of disconnected prompts.

Good fit when

  • You are starting a 1C project with Claude Code and want ready roles, skills and rules from the start
  • You need explicit control over the agent's permissions for each source folder
  • You are ready to connect the 1c_mcp MCP server to your base for checking against real metadata

Not a fit when

  • You do not have Windows or PowerShell: the template's batch scripts are built for them
  • You do not use Claude Code
  • You need tools for directly editing XML metadata structure: the template deliberately leaves those out

Example request

Run /1c-feature-dev to add a service desk API integration and take the task through the review and simplification subagents

Limitations

There is no license file in the repository. Full functionality requires installing the third-party 1c_mcp MCP server by Vladimir Kharin into the infobase and publishing its HTTP service; without it the agent loses access to metadata and data. Batch export and import of the configuration go through the real Configurator and can take noticeable time on large bases. The template has no tools for editing XML metadata structure directly, recommending the separate cc-1c-skills project for that.

How to disable. Remove the .claude folder and the CLAUDE.md file from the project, along with entries in .mcp.json if MCP servers were connected.

Security check

  • Batch scripts can export and import the configuration through the real Configurator
  • Agent permissions are configured by the user and cautious by default, but can be widened to direct writes to source files

README in short

The README describes the template's features (BSL norms, MCP as source of truth, 16 skills, 6 subagents, permission control, batch operations), a step-by-step quick start via /setup-project, updating the template, working on tasks in full and light cycles, tables of all skills and subagents, the WHAT/HOW separation principle, the connected MCP servers (1c_mcp and v8std), and credits to the source projects: 1c_mcp, v8std.ru and 1c-ai-feature-dev-workflow.

FAQ

Does the template work without MCP?

Yes, but the agent relies only on source files and honestly flags that it could not check against the live base.

How do you update the template in an existing project?

Copy the .claude folder over your own, leaving CLAUDE.md, .mcp.json and other already configured root files untouched.

Editors’ pick

A skills library that gives coding agents a development process: brainstorming, planning, TDD, subagents and code review

PluginMedium riskNo VPN needed292.5KRepository stars
Editors’ pick

Small composable skills for engineering with agents: plan grilling, TDD, bug diagnosis, code review and architecture

SkillLow risk271.4KRepository stars
Editors’ pick

GitHub toolkit for spec-driven development: the specify CLI adds agent commands and skills to a project, from principles to implementation

CLIMedium riskNo VPN needed139.3KRepository stars

Reference MCP servers

Model Context Protocol servers

Official

Official reference MCP servers: Filesystem, Fetch, Git, Memory, Sequential Thinking, Time and Everything

MCP serverMedium risk90.6KRepository stars
Foxx AI1C AI DevStart

I am Foxx AI and I have already vetted this tool. Ask about install, setup or anything else, and I will keep it simple.