Полная документация ниже — автоматически из репозитория svjszn/bitrix24-openlines-skill (ветка master). Оригинальные материалы проекта распространяются на условиях его лицензии.

bitrix24-open-lines-bot

A Claude Code Skill with empirically-verified knowledge for building a chatbot that answers inside an existing Bitrix24 Open Line (Открытая линия) — registration via imbot.v2, the incoming event contract, replying, and handing off to a human operator.

This isn’t a bot framework — it’s a skill file (SKILL.md) that an AI coding agent loads into context before writing or debugging a Bitrix24 Open Lines integration, capturing the parts of this feature that are either undocumented, easy to misconfigure, or actively misleading if you go by official tutorials alone.

Why

Bitrix24’s chatbot/Open Lines stack has more than one way to look correctly configured while silently not working (see "the WELCOME_BOT_ID trap" below). Most of what’s in this skill was confirmed against a real portal — sending real messages through a real connector and observing what actually arrives — not just copied from docs. Findings are dated so you can judge whether Bitrix24’s behavior may have since changed.

What’s covered

• The two integration paths and which one you actually want: a bot inside an existing Open Line (imbot.v2.Bot.register) vs. registering a brand-new channel/connector type (imconnector.*) — and why the latter flatly rejects webhook auth (WRONG_AUTH_TYPE, OAuth-app-only)
• Why imopenlines.config.update with WELCOME_BOT_ID/WELCOME_BOT_ENABLE looks like it attaches your bot to the line but does not — that’s a separate, older "queue greeting bot" feature that never triggers ONIMBOTV2MESSAGEADD
• The bot registration call, its real (nested, not top-level) response shape, and how to make it idempotent across restarts
• Read-only, zero-side-effect ways to sanity-check a webhook’s actual permissions before writing anything
• The incoming event contract (ONIMBOTV2MESSAGEADD), including the only authenticity check available (auth.application_token) and how to filter to Open Lines chats specifically
• Replying (imopenlines.bot.session.message.send) and handing off/closing a session (.operator, .transfer, .finish)
• A generic pattern for wiring this into an agent/LLM-driven conversation loop

See skills/bitrix24-open-lines-bot/SKILL.md for the full content.

Related skill

This skill assumes bitrix24-rest-api for general auth/retry/error-shape conventions — load that one first.

Using this skill

The skill itself lives in skills/bitrix24-open-lines-bot/.

Claude Code: copy or symlink that directory into your skills path (e.g. .claude/skills/bitrix24-open-lines-bot/ for a project-local skill, or the equivalent global skills directory), or clone this repo and point at it directly:

git clone https://github.com/soajv/bitrix24-openlines-skill.git
cp -r bitrix24-openlines-skill/skills/bitrix24-open-lines-bot .claude/skills/bitrix24-open-lines-bot

Other agents/tools: skills/bitrix24-open-lines-bot/SKILL.md is plain Markdown with a YAML frontmatter header (name, description) — feed it into any system prompt or context-loading mechanism that accepts freeform reference material.

Contributing

Corrections and additional confirmed gotchas are welcome. Please:

• Verify claims against a real portal where possible (ideally with a real connector, not just the API), and say what you tested and when.
• Cite the specific official docs page you’re relying on, if any.
• Keep the tone of existing entries: state the behavior, the evidence for it, and how to work around it — not just a link.

Disclaimer

This project is unaffiliated with, and not endorsed by, Bitrix24 or 1C-Bitrix. "Bitrix24" is a trademark of its respective owner. Information here reflects behavior observed as of the dates noted in the text and may not be current — always cross-check against official docs for anything security- or billing-sensitive.

License

MIT