Mcp Server
Give your agent a single MCP entry point to run AgentLed workflows with memory and many third-party integrations under one API key and credit pool.
Overview
AgentLed MCP is an MCP server for the Build phase that orchestrates AI workflows with long-term memory, 100+ integrations, and unified workspace credits via a stdio npm package.
What is this MCP server?
- stdio npm package @agentled/mcp-server (manifest version 0.7.5)
- Workflow orchestration with long-term memory across sessions
- 100+ integrations exposed through AgentLed’s unified platform
- Unified credits model via AGENTLED_API_KEY from Workspace Settings → Developer
- Package @agentled/mcp-server version 0.7.5
- 100+ integrations cited in server description
- Required secret env: AGENTLED_API_KEY
Community signal: 2 GitHub stars.
What problem does it solve?
Your agent hits a wall when real tasks need remembered context and multiple SaaS APIs, and you do not want to maintain a separate MCP server per integration.
Who is it for?
Builders already on AgentLed who want Claude Code or Cursor to drive multi-step automations with memory and a broad integration catalog.
Skip if: Developers who only need one fixed API (use a narrow MCP) or who cannot use a hosted orchestration platform with API keys and credits.
What do I get? / Deliverables
After you set AGENTLED_API_KEY and register the stdio server, your agent can run AgentLed workflows and integrations from one MCP connection.
- Local stdio MCP bridge to AgentLed orchestration APIs
- Agent-invokable workflows across 100+ platform integrations with memory
Recommended MCP Servers
Journey fit
While you are assembling the product and agent stack, this server belongs in Build because it extends what your coding agent can orchestrate—not a one-off launch or ops monitor. Agent-tooling is the canonical shelf for MCP servers that coordinate multi-step AI workflows, integrations, and persistent context for builders.
How it compares
Agent workflow orchestration MCP with platform integrations, not a single-purpose GitHub or Postgres MCP.
Common Questions / FAQ
Who is AgentLed MCP for?
Solo builders and small teams using AgentLed workspaces who want their coding agent to orchestrate workflows and integrations through MCP.
When should I use AgentLed MCP?
Use it during Build when you are wiring agent-tooling and need chained tasks, persisted memory, and many connectors without writing each MCP yourself.
How do I add AgentLed MCP to my agent?
Install @agentled/mcp-server from npm, set AGENTLED_API_KEY from Workspace Settings → Developer, and register the stdio MCP server in Claude Code, Cursor, or your client’s config.