Workspace MCP: Advanced
Rate limits, multi-client setup, troubleshooting, and security guidance for running the Taskade Workspace MCP server in production.
This page covers what comes after the basic setup in Workspace MCP — rate limits, running multiple clients at once, the differences between inbound and outbound MCP, troubleshooting, and security.
Table of Contents
Inbound vs Outbound MCP
Taskade uses MCP in two directions. Knowing which is which unlocks the right integration pattern.
Inbound
AI tools like Claude Desktop
"Claude, list my Taskade projects"
Workspace MCP (this page adds advanced config)
Authentication & Token Scoping
The inbound MCP server (@taskade/mcp-server) authenticates with a Personal Access Token via the TASKADE_API_KEY environment variable.
Token best practices
Scope narrowly. Create a dedicated token per workspace or per use case rather than reusing one token everywhere.
Rotate every 90 days. Regenerate and update all client configs.
Never commit tokens to version control or share them in chat.
Revoke unused tokens from taskade.com/settings/api.
OAuth availability
OAuth 2.0 is available for the Genesis App MCP (which runs hosted at a URL). The local @taskade/mcp-server inbound server currently uses personal tokens only.
Rate Limits
MCP requests share the Taskade API rate limit budget.
Tool call returns 429
Rate limit exceeded
Implement backoff in your client wrapper
Multiple tools failing simultaneously
Token-wide rate limit exhausted
Reduce concurrency; split tokens
Slow tool response
Upstream model slowness
Check model pricing tier; auto-mode routes dynamically
If you hit limits regularly, split your integration across multiple scoped tokens so each has its own budget.
Multi-Client Setup
You can safely run @taskade/mcp-server on Claude Desktop, Cursor, Claude Code, Windsurf, VS Code, and other clients at the same time. Each client spawns its own stdio process — state is isolated server-side per token.
Claude Desktop
File: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS)
Cursor
File: .cursor/mcp.json in your project root.
Claude Code
Windsurf
File: ~/.codeium/windsurf/mcp_config.json (or Settings → Cascade → MCP → Add Server):
VS Code
Add a .vscode/mcp.json to your workspace (VS Code uses the servers key; ${input:…} prompts for your key on first run):
Avoid committing these config files to public repos with real tokens. Many teams use environment variable substitution (shell-level) or a secret manager to inject the token at runtime.
Plan Gating
Inbound MCP (@taskade/mcp-server)
Limited
Limited
Full
Full
MCP Connectors (outbound)
—
—
✓
✓
Taskade-as-MCP-server (external clients read workspace)
—
—
✓
✓
Custom domain for MCP
—
—
—
✓
Plan features may evolve. Check the pricing page for current gating.
Tool Catalog Details
The inbound server exposes 62 tools across 8 categories. Most mirror the REST API v1 operations it wraps; the newest category is a small Agent Chat & Webhooks (API v2, beta) group (promptAgent, listConversations, getConversation, subscribeWebhook, unsubscribeWebhook). Below are the ones integrators most often need to configure precisely.
projectTasksGet
projectTasksGetRequired args:
projectIdOptional args:
limit(default 100),after,beforePagination: Cursor-based — pass the last task id as
afterto page forward.
taskCreate
taskCreateRequired args:
projectId,tasks(array of{ contentType, content })Optional args:
placement(afterbegin|beforeend)
agentConvosGet / agentConvoGet
agentConvosGet / agentConvoGetagentConvosGetlists an agent's conversations (agentId, optionallimit,page).agentConvoGetreturns one conversation (agentId,convoId).
This server now bundles a small API v2 (beta) layer (prompt-an-agent promptAgent, agent-chat, webhook subscribe/unsubscribe) on top of the v1 surface. Note: bundle export/import still lives in the Action API v2, not in this server.
For the full tool list, see the Workspace MCP reference.
MCP Connectors
MCP Connectors let your Taskade agents connect outbound to third-party services. You browse and enable connectors from the Integrations screen in your workspace — no code, no hosting.

Agents see enabled connectors as tools. You can opt specific tools out per agent (especially useful for public-facing agents). See MCP Connectors for the full catalog and setup steps.
Troubleshooting
"Connection refused" in Claude Desktop
MCP process crashed
Restart Claude Desktop; check ~/Library/Logs/Claude/
"Unauthorized" on every tool
Token invalid or rotated
Regenerate token; update all client configs
"Workspace not found"
Token scoped to wrong workspace
Create a token in the right workspace
Tools appear but return 429
Rate limited
Back off; consider splitting tokens
Agent invisible in shared workspace
Permission issue (fixed v6.114.1)
Update to latest @taskade/mcp-server
OAuth loop (Genesis App MCP)
Expired refresh token
Re-authenticate in the client
Tool timeout
Large response or slow upstream
Check upstream; reduce query scope
Still stuck? File an issue at github.com/taskade/taskade/issues with MCP logs.
Security Best Practices
Audit tool exposure for public agents. Opt sensitive tools (file access, automation triggers) out of public agent configurations.
Use per-workspace tokens. A leaked personal token is bounded to one workspace if scoped correctly.
Rotate on every personnel change. When a teammate leaves, rotate any shared tokens.
Monitor the workspace activity log for unexpected MCP-initiated actions.
Use TLS / custom domain for Genesis App MCP in production.
Related
Workspace MCPHosted MCP — Genesis App (Beta)Action API v2 ReferenceLast updated
Was this helpful?