bsl-ai-toolkit skill
A set of skills, rules and MCP servers for 1C Enterprise 8.3 development where the agent works from real code and docs, not from memory
- 35 ★ on the repository
- Last commit
- MIT
Broad access
Why this level
- Edits metadata and configuration code with apply into a test database
- Reaches into database data and operations systems
- Uses ssh, scp and credentials via environment variables
Install
- Works in
- Services
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/bsl-ai-toolkitDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Edits metadata and configuration code with apply into a test database
- Reaches into database data and operations systems
- Uses ssh, scp and credentials via environment variables
This is third-party code. Review the repository files before installing.
What it does
A reusable toolkit for 1C Enterprise 8.3 development in the BSL language: rules, eight role-based skills, MCP servers and a deterministic metadata editing engine. A single core in the core directory is shared across agents, and thin adapters wire up Claude Code, Cursor, Copilot, Gemini CLI, Codex, Cline and Windsurf. The MCP servers give search over the real configuration and extension code, BSL Language Server diagnostics after an edit, read of operations data and read-only access to database data. The engine edits forms, data composition schemas and attributes in the Configurator and EDT formats with a round-trip check, so the model does not invent methods or structure.
Who it is for. For 1C developers and teams who need verifiable AI-assisted development from real code rather than the model's guesses.
Good fit when
- You write or review BSL code and want checks against real code via MCP
- You need to edit forms, data composition schemas or attributes without breaking the format by hand
- You run 1C development across several agents and operating systems
Not a fit when
- The task is not about 1C Enterprise and the BSL language
- You have no access to the configuration code, platform or test database the toolkit relies on
Example request
Add an attribute to the document through the metadata engine and check BSL diagnostics after the editLimitations
The metadata engine is pure Python, but full value needs local clones of the configuration and extension code, a 1C platform and a test database, and the apply layer uses ssh, scp and tar or a local runner. Some MCP servers require external tools: the BSL Language Server, a JDK, and optionally Zabbix or Prometheus. Paths and credentials are set through environment variables. Secrets and personal data are not sent to the model, and database data is read only, with masking.
How to disable. Remove the generated toolkit files from the project: CLAUDE.md, AGENTS.md, GEMINI.md, .mcp.json and the .claude directory, and delete the repository clone.
FAQ
Is the toolkit tied to one agent?
No. The core is shared, and adapters wire up Claude Code, Cursor, Copilot, Gemini CLI, Codex, Cline and Windsurf. Claude Code is the reference implementation.
What does byte-perfect metadata editing give?
Changes to forms, data composition schemas and attributes are deterministic and verified by a round-trip into a test database, so the diff is exactly the target and layout does not break.
Discussion
No questions or reviews yet.
Related
All 1C:Enterprise tools1C AI development environment
1C AI Development Environment
1C development environment with an MCP server offering 10 tools: 11 BSL analyzers, JSON DSL to XML, dependency graph, and EPF build without 1C
1C AI Development Kit: 1C skill collection
1C AI Development Kit
A set of 52 Claude Code skills and agents, plus Cursor rules, for the full 1C:Enterprise 8.3 development cycle, assembled from several community projects
Rules, subagents, skills and MCP integrations for 1C:Enterprise 8 development with Cursor, Claude Code, Codex and other agents
MCP server with the 1C platform reference and validation of BSL code against the real API of a chosen platform version
claude_rules_1c: Claude Code rules and agents for 1C
claude_rules_1c
A set of 12 subagents, profile-based rules and MCP configuration for 1C development in Claude Code, with a three-layer rule loading system
A kit of skills, agents and rules for 1C development in Claude Code and Cursor: metadata, forms, DCS, extensions, infobases and web client
README in short
The README presents bsl-ai-toolkit as a reusable toolkit for AI-assisted 1C Enterprise 8.3 development: rules, eight skills, MCP servers and a metadata editing engine. The idea is verifiability through tools: onec-code search over real code, bsl-ls diagnostics after an edit, the deterministic onec_metadata engine with round-trip, onec-ops for operations and read-only onec-data for database data. The core directory is shared across agents, thin adapters wire up specific hosts, build.sh assembles configs and onboard runs onboarding. It claims cross-platform support and operation under limited model access.
README badge
Original SKILL.md
--- name: 1c-dev description: > Разработка и код-ревью на 1С:Предприятие 8.3 (BSL) с проверкой по РЕАЛЬНОМУ коду через MCP, а не по памяти. Используй всякий раз, когда пишешь, правишь или ревьюишь BSL-код, работаешь с общими модулями, формами, запросами, СКД, регистрами, ролями; реализуешь задачу по ЧТЗ/тех-проекту; ищешь, как устроена типовая конфигурация или что уже есть в расширении. Железное правило: НЕ выдумывай поведение 1С и сигнатуры — проверяй по реальному коду через MCP и справку платформы, после правки прогоняй BSL Language Server. Анализ/архитектура — скилл 1c-analyst. --- # Разработка 1С (BSL) — писать корректный и производительный код по реальному коду ## Главное правило: спроси инструмент, не угадывай Нейросеть врёт в деталях 1С. Источник истины — реальный код и официальная справка.