1C feature development plugin for Claude Code
1C:Enterprise AI Feature Development Plugin
A Claude Code plugin with a seven-phase workflow for extending 1C configurations: code research, architecture, implementation and review through separate agents
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
- The 1c-code-writer subagent writes code directly into 1C configuration files
- The workflow depends on an external MCP server that validates the changed objects
Install
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/1c-enterprise-ai-feature-development-pluginDetects the agents on your machine, checks the risk and pins the version.
Other ways to install
Run one by one in the Claude Code chat
/plugin marketplace add diversus23/1c-feature-dev
/plugin install 1c-feature-dev@1c-feature-dev-marketplaceRun /plugin marketplace add Diversus23/1c-feature-dev then /plugin install 1c-feature-dev, make sure the EDT MCP Server from https://github.com/DitriXNew/EDT-MCP is connected, and start /1c-feature-dev with your task description.
Other ways from the author
/plugin marketplace add Diversus23/1c-feature-dev
/plugin install 1c-feature-devThe README's recommended method.
This is third-party code. Review the repository files before installing.
What it does
The plugin adds a /1c-feature-dev command that walks a task through seven phases: checking MCP server availability, gathering requirements with a complexity estimate, exploring the codebase, designing architecture, implementation, review and a summary. Each phase runs a dedicated subagent (explorer, architect, code writer, reviewer, simplifier, test generator, documenter), and how many run in parallel depends on task complexity. Separate commands cover /1c-query-optimization for query optimization patterns and /1c-docs for documentation generation. Coding rules, working with the standard subsystems library and the EDT metadata format load automatically from the rules folder.
Who it is for. For 1C developers who extend EDT-format configurations through Claude Code and want a structured process instead of one-off prompts.
Good fit when
- You want to run a 1C configuration change through defined stages, from code research to review
- You need separate subagents for 1C architecture, code writing and review
- You need YaXUnit tests, Vanessa Automation scenarios or technical documentation generated from the change
Not a fit when
- The configuration is not exported in EDT format
- You need a quick one-off edit rather than the full phased workflow
- You cannot run the EDT MCP Server that the workflow checks for at startup
Example request
/1c-feature-dev add integration with the warehouse system APILimitations
The workflow needs a 1C configuration exported in the 1C:EDT format and a separately installed EDT MCP Server for syntax checks and lookups against the standard library and metadata: without it, phase 0 stops the run. The project is built on top of three third-party works (cursor_rules_1c, Anthropic's feature-dev, 1c-feature-dev-workflow), and their licensing terms are not automatically carried over to the plugin repository, so that is worth checking separately. Code implementation only starts after the user explicitly approves the plan.
How to disable. Run /plugin uninstall 1c-feature-dev in Claude Code, or, if installed manually, delete the copied files from ~/.claude/agents, ~/.claude/skills and ~/.claude/rules.
Security check
- The 1c-code-writer subagent writes code directly into 1C configuration files
- The workflow depends on an external MCP server that validates the changed objects
README in short
The README describes a Claude Code plugin for extending 1C configurations through a seven-phase numbered workflow, with a mermaid diagram, a phase table, a command list, an agent list with the model used (sonnet or opus) and its role. It separately lists auto-loaded coding rules and two install paths: through the Claude Code plugin marketplace and by copying files manually. Requirements call for an EDT-format configuration export and an external EDT MCP Server. At the end it credits the third-party projects the plugin builds on.
FAQ
Can the plugin's agents be used separately from the full workflow?
Yes, the README states that agents like 1c-code-explorer, 1c-code-architect and 1c-code-writer can be called on their own.
What does phase 0 do?
It checks that the required MCP servers are reachable. If not, the workflow stops until they are connected.
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