- MONITOR
- INSPECT
- TERMINATE
- HUMAN
Short answer
How can I track and stop my Lindy agent runs?
You can pull a list of all current tasks to see what's active, check the specific details of a single run, or stop a process immediately if it's looping. You'll see how to manage these states below.
Control your agent's activity.
Manage your Lindy workforce.
Use these outcomes to keep your autonomous employees on track.
MONITOR
Track active tasks
The AI pulls a list of all current runs so you know exactly what's happening in the background.
INSPECT
Check specific run details
Get a deep dive into a single task to see its current status and progress.
TERMINATE
Stop runaway processes
Kill a specific run instantly if the agent enters a logic loop or starts behaving unexpectedly.
HUMAN
Manual intervention
You decide when a task is finished or when it needs to be stopped, keeping the AI on track.
The workflow
What your AI does when <the thing arrives>.
Your agent handles the heavy lifting of checking statuses and executing stops.
Identify active tasks
The AI scans your Lindy account to find every task currently in progress.
list_runsVerify task status
Once you find a specific run, the AI looks up its exact state and current progress.
get_runStop a loop
If a task is stuck or wasting resources, the AI sends a command to kill that specific run.
cancel_runConfirm termination
The AI checks the run status again to ensure the process actually stopped.
get_run
Try it
Copy these to start.
Use these prompts to take control of your Lindy agents.
Starting points
Swap out the specific run IDs or task names for the ones you actually want to manage.
Lindy Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_c2F5Hk7TPFg1Cz9vjMKKvEPPvQH0HYkK2vL7pHhF/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Lindy capabilities are ready to use.
{
"mcpServers": {
"lindy-autonomous-ai-employees-mcp": {
"url": "https://edge.vinkius.com/vk_preview_c2F5Hk7TPFg1Cz9vjMKKvEPPvQH0HYkK2vL7pHhF/mcp"
}
}
}Show me all the Lindy agent runs currently in progress.
Check the status of the Lindy run with ID run_98765.
Stop the Lindy agent run for the customer outreach task immediately.
Is the Lindy run for the invoice processing still active?
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Lindy once, then ask.
Just link your account once. Your credentials stay encrypted, and you can start managing runs through any compatible client immediately.
Connect Lindy to your AIFAQ
How the monitoring works
- 01
Can the AI delete my agent's history?
No. The tools only allow for listing, getting, and canceling runs. It cannot delete data or history.
- 02
Can I use this to change the agent's instructions?
No. This MCP is strictly for monitoring and stopping runs. It cannot modify how an agent is configured.
- 03
How do I stop an agent that is stuck in a loop?
You can tell your AI to cancel the specific run using the cancel tool.
- 04
Can the AI see all my Lindy data?
The AI only sees the run data provided by the specific tools you authorize it to use.
- 05
Will the AI automatically stop a run if it takes too long?
No. You must tell the AI to check the status and then decide to cancel the run yourself.
More questions about Lindy? The Connector page answers them. See everything the Lindy Connector can do
Connect Lindy to Claude, Cursor, ChatGPT & more
