Runbear · Works with Claude
Claude is your AI assistant.
Runbear is your team's AI teammate.
Claude lives in a chat window.
Runbear lives in your Slack, with its own identity, knowledge, and permissions. Your team gets named AI teammates that speak up when they should, not just when someone asks.
Why not just Claude
Claude is personal.
Runbear is organizational.
Claude is the best AI assistant in the world for one person, one session, one question.
Runbear is the layer on top, so your whole company can work with AI the way a team works together: shared identity, shared knowledge, shared permissions.
- Personal. One user, one session, one Anthropic account.
- Reactive. Answers when you ask. Waits for the prompt.
- Per seat. Every user who needs AI needs a Claude Team seat.
- One identity. @Claude, one bot for the whole workspace.
- Organizational. Shared agents with team knowledge and roles.
- Proactive. Watches channels, monitors triggers, speaks up when it should.
- Per usage. Pay for what your team actually runs, not for idle seats.
- Many identities. @HR, @Platform, @SalesOps, each with its own permissions.
Reactive vs. proactive
Claude responds when you ask.
Runbear speaks up when it should.
Most AI only works if someone remembers to open it. Runbear agents are always on. They watch the channels they belong to, trigger on events, and prepare answers before the question even lands.
migration.user_events. This looks like incident #412. Rollback PR is drafted. Paging @maya (table owner). I'll open a ticket if no one replies in 5 minutes.One bot isn't a team
Claude gives you @Claude.
Runbear gives you a roster.
One assistant can't be your HR, your platform oncall, and your sales ops at the same time. Runbear lets you stand up named agents, each with its own role, its own knowledge base, and its own permissions.
One identity for everything.
HR question? @Claude. Deploy issue? @Claude. Deal question? @Claude.
Claude + Runbear
Think in Claude. Ship with Runbear.
Every Claude conversation you have privately can become a Runbear agent your whole team shares. Claude is where you think. Runbear is where your team acts.
Prototype the prompt.
Use Claude.ai or Claude Code to iterate on a system prompt, a workflow, or a decision tree until it works for you. One person, one conversation, fast feedback.
Wrap it in a teammate.
Bring the prompt into Runbear. Attach team knowledge, tool connections, and channel triggers. Give the agent a handle, a home channel, and permissions.
Your team uses it every day.
@the-agent answers in-channel, acts on triggers, and respects each person's access. The workflow you discovered once is now part of how the whole team works.
Questions you're probably asking
Runbear and Claude, side by side.
Give your team an AI teammate, not another AI tool.
