By runbear-io
Deploy a local Claude Code project to a hosted Runbear Agent SDK agent, connect it to Slack, and manage its event triggers and scheduled jobs
Connect a Runbear Claude Agent SDK agent to Slack through the Runbear management MCP (`deploy_to_slack`). Defaults to a workspace-specific custom Slack bot (`mode: custom`) — the agent gets its own bot identity instead of the shared @Runbear bot — and can join public channels afterward. Use when the user wants to connect, deploy, install, or publish a Runbear agent to Slack as its own bot, optionally joining channels. The target agent is given by app UUID, agent URL, or agent name (fuzzy-matched).
Deploy a local Claude Code project to a Runbear Claude Agent SDK agent AND configure the MCPs and skills it needs, so colleagues can use your agent without you doing it on their behalf. Files are uploaded through the Runbear signed-URL flow (`create_project_upload` + `finalize_project_upload`); MCPs are attached with a per-integration choice of shared vs per-user auth; skills are curated from your project and user scope; and the whole configuration is saved to a committed `.runbear/deploy.json` manifest for one-command redeploys. After deploying, if the agent has no Slack channel yet, it offers to connect Slack (via `runbear:connect-slack`) so colleagues can reach it. Use when the user wants to deploy, publish, push, sync, share, or redeploy a local project (its files, CLAUDE.md, skills, MCPs, and code) to a hosted Runbear Agent SDK agent identified by an app UUID, agent URL, or agent name (fuzzy-matched) — or, with `--new <agentName>`, to create a brand-new Claude Agent SDK agent and deploy to it in one step.
Set up and manage a Runbear agent's event triggers and scheduled jobs through the Runbear management MCP — event triggers (Pipedream-driven, fire when something happens in an external app like a new Gmail, a new Notion page, or a GitHub issue) and scheduled jobs (run the agent on a cron schedule or once at a set time). Use when the user wants to "trigger the agent when X happens", "run the agent every morning / weekly / on a schedule", "remind me / run this once at <time>", "notify me on Slack or email when <event>", or to list, inspect, pause, resume, or delete an agent's existing triggers. Standalone: it owns the cross-tool decision tree and preconditions; the individual MCP tools own their own mechanics. The target agent is given by app UUID, agent URL, agent name (fuzzy-matched), or "my personal agent".
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Skills for exposing Claude Agent via HTTP, Slack, and Discord.
/plugin marketplace add runbear-io/skills
/plugin install dispatch-http@runbear-skills
/plugin install dispatch-slack@runbear-skills
/plugin install dispatch-discord@runbear-skills
/plugin install skill-uploader@runbear-skills
/plugin install runbear@runbear-skills
Expose Claude Code as REST API endpoints.
| Skill | Description |
|---|---|
dispatch | Start an Express server that exposes Claude Code/Cowork via REST endpoints |
expose | Expose the local dispatch-http server to the internet via a Cloudflare quick tunnel |
/dispatch-http:dispatch
/dispatch-http:dispatch port 8080
/dispatch-http:expose
Connect Claude Code to Slack via Socket Mode. Supports streaming responses.
| Skill | Description |
|---|---|
dispatch | Start/stop the Slack bot server that connects Claude Code to Slack |
init | Create and configure a Slack bot app with OAuth token rotation |
/dispatch-slack:init
/dispatch-slack:dispatch [start|stop|status] [/path/to/project]
The dispatch skill supports two modes:
CLAUDE_CODE_OAUTH_TOKEN (generate with claude setup-token) or ANTHROPIC_API_KEY in .env.Connect Claude Code to Discord. Supports streaming responses via progressive message editing.
| Skill | Description |
|---|---|
dispatch | Start/stop the Discord bot server that connects Claude Code to Discord |
init | Create and configure a Discord bot via the Developer Portal |
/dispatch-discord:init
/dispatch-discord:dispatch [start|stop|status] [/path/to/project]
The dispatch skill supports two modes:
CLAUDE_CODE_OAUTH_TOKEN (generate with claude setup-token) or ANTHROPIC_API_KEY in .env.Deploy local Claude skill folder(s) to a hosted Runbear Agent SDK agent. Reads a local
skill folder, runs a safety preflight, and deploys each SKILL.md through the Runbear
upload_local_skill_to_agent MCP tool.
| Skill | Description |
|---|---|
upload | Upload local skill folder(s) to a Runbear agent identified by an app URL or app UUID |
upload_local_skill_to_agent tool.--agent)./skill-uploader:upload ./my-skill --agent https://app.runbear.io/agents/<appId>
/skill-uploader:upload ./my-skill --agent <appId>
/skill-uploader:upload ./my-skill --agent <appId> --overwrite
/skill-uploader:upload ./internal-skills --agent <appId>
The backend deploys one skill per call. If the folder contains multiple SKILL.md
files, each is uploaded independently and reported per skill.
Deploy a whole local Claude Code project to a hosted Runbear Agent SDK agent and
configure the MCPs and skills it needs, so colleagues can use your agent without you
doing it on their behalf. The hosted agent runs against the same CLAUDE.md, skills,
subagents, docs, and code you have locally, with the directory layout preserved.
| Skill | Description |
|---|---|
deploy | Deploy a local project's files, select which local MCPs the agent gets (shared vs per-user auth per MCP), and curate its skills — all saved to a committed manifest for one-command redeploys |
connect-slack | Connect a Runbear Agent SDK agent to Slack — creates a workspace-specific custom bot by default, and can join channels |
deploy configuresnpx claudepluginhub runbear-io/skills --plugin runbearClaude Code plugin for managing multiple AI agents on scheduled weekly plans. Each agent is a 1-to-1 wrapper around a Claude Code subagent with its own long-term goals, monthly objectives, weekly tasks, and token budget.
BearDrive — the open-source Google Drive for AI agents: mount folders that stay in sync across devices and teammates through a self-hostable BearDrive hub (backed by S3/GCS/any object store), with turn-boundary sync, per-change attribution, and full change history. CLI: bdrive.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
This skill should be used when users need to generate ideas, explore creative solutions, or systematically brainstorm approaches to problems. Use when users request help with ideation, content planning, product features, marketing campaigns, strategic planning, creative writing, or any task requiring structured idea generation. The skill provides 30+ research-validated prompt patterns across 14 categories with exact templates, success metrics, and domain-specific applications.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Consult multiple AI coding agents (Gemini, OpenAI, Grok, Perplexity, plus codex, antigravity, and grok CLIs when installed) to get diverse perspectives on coding problems