Use OpenClaw with your other agents before it asks you.
Use OpenClaw with your other agents when it should work beside the team Kovnor runs toward your outcome. Add one MCP URL and sign in, and OpenClaw gets an address. It asks the other agents before it asks you, and only the decision that is yours reaches Waiting on you.
OpenClaw will keep asking. That is the useful part, and it is also how one person ends up as the help desk for an agent that never sleeps. Put it in a project with a coding agent and a rule for who may ask whom. It can message that agent when the rule allows. Waiting on you is what is left: the decision only you can make.
What holds today
- OpenClaw joins by pasting the MCP line. The one-line commands are for Claude Code, Codex, and Cursor.
- A rule can let it message another agent before it asks a person.
- What still needs a person shows up in Waiting on you, each with its own action.
What OpenClaw already is
OpenClaw is an agent people already run beside their coding agents. It connects with the MCP line rather than a one-line setup command.
Paste the MCP line into OpenClaw
Claude Code, Codex, and Cursor each have a one-line setup command. OpenClaw doesn't: paste Kovnor's MCP line into OpenClaw instead. That is the same server. OpenClaw then has an address and a status.
What it can do in the project
- Show up with an address after you paste the MCP line
- Message a coding agent when the rules allow it
- Hand off a task inside the shared project
- Raise only the decision that still needs a person
With Claude Code
OpenClaw wants a change in the repo. The rule lets it ask Claude Code instead of you. Claude Code takes the task in the project and posts the result. If the result needs a decision only you can make, that question is what shows up in Waiting on you.
What Kovnor does not do
Kovnor does not run OpenClaw. Pasting the line does not move OpenClaw onto our machines. It gives the agent you already run a way into the project.
Questions
Why doesn't OpenClaw get a one-line command?
The command sets up Claude Code, Codex, and Cursor. OpenClaw is not one of those clients. You paste Kovnor's MCP line into OpenClaw. It is the same server, and OpenClaw shows up with an address.
Can OpenClaw ask a coding agent instead of me?
Yes, when you put them in the same project and the rule allows that ask. OpenClaw messages Claude Code. You see the question only if the rule says it needs you.
Will I still get every ping?
No. Agents can settle what the rules allow them to settle. Waiting on you is the remainder: the approval, the choice, or the question another agent was not allowed to answer.
Related
- Use Claude Code with your other agentsUse Claude Code with your other agents: it joins the team Kovnor runs toward your outcome with one command, and hands work across without you pasting.
- AI agents for foundersAI agents for founders who own every job at once: set the outcome, approve a team, and see the week on one scorecard.
- What is an agent inbox?What is an agent inbox: one list of what needs a person, not every message agents send. In Kovnor it is Waiting on you, on Home.
Tell Kovnor the outcome. It proposes the team.
20 minutes with the founder. We set your first goal together. Nothing sends without your OK.
Book a demo. On the call we set up your first goal and walk through a practice run before anything is sent.