# Lindy (Autonomous AI Employees) MCP > Lindy (Autonomous AI Employees) MCP gives you a direct line to your Lindy.ai workspace from your favorite AI client. You can trigger autonomous agents, inspect their internal reasoning logs, monitor active task runs, and manage your entire fleet of Lindies without ever leaving your chat interface. It turns your agent into a command center for your autonomous workforce. ## Overview - **Category:** superpower - **Price:** Free - **Tags:** orchestration, agent-management, task-automation, audit-logs, workflow-monitoring, integration-management ## Description Managing a fleet of autonomous agents usually means jumping between browser tabs, hunting for specific run IDs, and squinting at logs to figure out why a task stalled. This MCP changes that by bringing your entire Lindy.ai workspace into your primary AI client. Instead of hunting for data, you just ask your agent what's happening. You can check the status of a specific run, see exactly what your agents are doing with your Slack or Gmail connections, and even pull the raw reasoning logs to see the internal logic of a loop. If an agent gets stuck in a context circle, you can kill the execution immediately via chat. It moves the control plane from a complex dashboard into a natural conversation. Because Vinkius lets you connect to the entire catalog of 4,000+ MCPs from one client, you can manage your Lindy workforce alongside your other business tools without switching apps. You get a high-level overview of your organizational boundaries and a granular way to trigger complex workflows using simple JSON payloads. It's about having one place to see, audit, and command every autonomous employee you've built. ## Tools ### cancel_run Interrupts a running execution and stops tasks stuck in context loops. ### get_run Gets the current state of a run that is blocked by an API or human input. ### get_lindy Retrieves the prompts and tool configurations for a specific Lindy. ### list_integrations Lists all third-party app connections like Slack and Gmail. ### list_runs Lists recent runs and validates the execution graph for active instances. ### list_lindies Lists all custom autonomous assistants built in your workspace. ### get_run_logs Dumps the literal LLM reasoning logs for a specific run loop. ### trigger_lindy Starts an asynchronous task run by parsing a JSON payload. ### list_triggers Shows how your agents are woken up by webhooks, cron jobs, or APIs. ### list_workspaces Lists the organizational boundaries and team structures in your account. ## Prompt Examples **Prompt:** ``` List all active Lindies in my workspace ``` **Response:** ``` I've found 3 autonomous assistants: 'Sales-Research-Lindy' (ID: l-123), 'Customer-Support-Lindy' (ID: l-456), and 'Content-Writer-Lindy' (ID: l-789). Which one would you like to trigger or inspect? ``` **Prompt:** ``` Show me the reasoning logs for the last run of 'Sales-Research-Lindy' ``` **Response:** ``` Retrieving logs for run ID 'run-98765'… I've extracted the reasoning steps. The Lindy searched for LinkedIn profiles, identified 5 targets, drafted personalized intro emails, and is now waiting for your approval to send them via Gmail. Would you like to see the draft email content? ``` **Prompt:** ``` What triggers are currently configured for our autonomous agents? ``` **Response:** ``` I've identified 3 active triggers: 1) Webhook (ID: tr-123) for 'Support-Lindy', 2) Scheduled Cron (Daily at 9 AM) for 'Report-Lindy', and 3) Manual API trigger for all assistants. I can provide the specific webhook URLs if needed. ``` ## Capabilities ### Trigger autonomous tasks Start new asynchronous runs for your Lindies using specific JSON payloads. ### Inspect agent reasoning Pull literal LLM reasoning logs to see how your agents are making decisions. ### Stop stuck executions Force a hard stop on running tasks that are caught in context loops. ### View workspace structures See how your Lindies are distributed across different team boundaries. ### Monitor active runs Track the real-time state of executions that are waiting on APIs or humans. ### Audit app connections See every third-party connection like Slack or Gmail linked to your AI. ## Use Cases ### Debugging a stuck support agent A support agent gets stuck in a loop. The developer uses get_run_logs to see the reasoning, identifies the error, and uses cancel_run to stop the loop. ### Auditing company integrations A founder wants to know which apps the AI can access. They ask the agent to list_integrations to see all active Slack and Gmail connections. ### Launching a research task An ops manager needs a specific research task done. They use trigger_lindy to pass a JSON payload to the Sales-Research-Lindy. ### Checking team boundaries A manager wants to see which assistants are assigned to which departments. They use list_workspaces to see the organizational structure. ## Benefits - Stop manual dashboard hopping by using list_runs to see every active agent status in one chat window. - Debug agent behavior instantly with get_run_logs to see the exact reasoning steps the AI took. - Kill runaway tasks immediately with cancel_run to prevent stuck loops from wasting your API credits. - Audit your security posture by using list_integrations to see every Slack and Gmail connection. - Launch complex workflows faster by using trigger_lindy to send JSON payloads directly from your chat. - Map out your entire AI infrastructure using list_workspaces to see how agents are distributed across teams. ## How It Works The bottom line is you get a chat-based command center for your autonomous agents. 1. Subscribe to the MCP and enter your Lindy API Token. 2. Connect your preferred AI client to the Vinkius catalog. 3. Ask your agent to list your Lindies or trigger a specific task run. ## Frequently Asked Questions **How do I stop a Lindy run that is stuck?** You can use the cancel_run tool to send a hard stop to any execution that is caught in a context loop. **Can I see the internal reasoning of my Lindy agents?** Yes, the get_run_logs tool allows you to dump the literal LLM reasoning logs for a specific run loop. **Does the Lindy MCP work with my Slack integrations?** You can use list_integrations to see all bounded third-party app connections, including Slack and Gmail. **How do I start a new task for a Lindy?** Use the trigger_lindy tool to start an asynchronous task run by passing in a JSON payload. **Can I see all my Lindies at once?** The list_lindies tool will enumerate every custom autonomous assistant built in your workspace. **How do I authenticate the Lindy MCP with my account?** You'll need a Lindy API Token. Grab that from your Lindy dashboard and paste it into your MCP client's configuration to connect your workspace. **Can I see what triggers my Lindy agents?** Yes, you can use the `list_triggers` tool. It shows how your agents get woken up, including Cron jobs, Webhooks, and manual API triggers. **How do I see how Lindies are organized across my company?** You can use `list_workspaces` to see your organizational boundaries. This helps you understand how your autonomous assistants are distributed across different teams. **Can I see exactly how my Lindy made a specific decision?** Yes. Use the `get_run_logs` tool with a specific Run ID. Your agent will retrieve the literal LLM reasoning loops and step-by-step validations, giving you full transparency into the autonomous agent's logic. **How do I trigger an autonomous task through a conversation?** The `trigger_lindy` tool allows you to start an asynchronous task run. You just need to provide the Lindy ID and a JSON payload defining the inputs for the task. Your agent will fire the job and return a Run ID for status tracking. **Can my agent list which third-party apps my Lindies are connected to?** Absolutely. Use the `list_integrations` tool to retrieve all active third-party app connections. Your agent will report which channels (like Slack, Gmail, or HubSpot) are securely connected to your workspace.