[feature]: Agent Mission Control + OpenClaw Integration (including $15/month Business plan and Agent Seat Cloning)

[speiron]

Is there an existing issue for this?

• I have searched the existing issues

Summary

Summary

I’d like Plane to support a native integration where Plane acts as a Mission Control for AI agents running on OpenClaw.

My use case is a solo operator managing 30+ specialized agents.
I need this feature to be available on the $15/month in Business plan (with fair limits), so independent builders can adopt it early and scale later.

Key Product Request

1) Agent as a First-Class Entity (based on Seat cloning)

Add an Agent entity that can be created by cloning a standard Seat/User profile structure, but typed as agent:

• seat_type: human | agent
• cloned permission/profile baseline from a Seat template
• agent-specific fields:
  • role/function
  • model profile
  • tool scope
  • automation policy
  • execution limits

This allows teams to reuse Plane’s existing seat architecture while clearly distinguishing machine actors from humans.

2) Availability in the $15/month plan

Please include an Agent Mission Control starter tier in the $15/month plan, for solo operators and small teams.

Suggested limits for this plan (example):

• up to 50 active agents
• limited run history retention
• basic mission-control dashboard
• webhook + callback support
• approval gates enabled

This is critical for adoption: many early users are solo builders with high automation needs but limited budget.

Why this matters

Without affordable entry pricing and native agent identity support, users must build custom infrastructure for features that should be platform-native.

With this, Plane can become the default operational layer for AI-native workflows.

Suggested MVP

• Agent Registry (Seat-clone model with seat_type=agent)
• Assign issues to agents
• Signed webhook trigger + callback status update
• Execution log in issue activity
• Idempotency key support
• Basic mission-control view
• Included in $15/month in Busines plan with starter limits

Acceptance Criteria

• I can create an Agent by cloning a Seat profile and marking it as agent.
• Agent permissions/scopes are configurable independently from human users.
• I can assign issues to agents and trigger external runs.
• I can enforce approval gates before sensitive actions.
• Duplicate events don’t create duplicate executions.
• Core Agent Mission Control is available in the $15/month in Business plan (with defined limits).

---

Why should this be worked on?

OpenClaw Integration

[makeplane]

Synced with Plane Workspace 🔄

[GIT-99] [feature]: Agent Mission Control + OpenClaw Integration (including $15/month Business plan and Agent Seat Cloning)

[Prashant-Surya]

Hey @speiron, thanks for the detailed feature request!

A lot of what you're describing is already possible with Plane Agents (currently in Beta). Agents in Plane are first-class workspace entities that can:

• Be @mentioned in work item comments to trigger interactions
• Be assigned work items directly, just like any team member
• Receive full work item, project, and conversation context via webhooks
• Track execution through the Agent Run lifecycle — with states like in_progress, awaiting, completed, failed, etc.
• Support multi-turn conversations with back-and-forth between users and agents
• Send thoughts (progress updates), actions (tool usage), responses, and elicitations (requesting user input) back to users
• Agents don't count as billable seats

You can check out the full documentation here: https://developers.plane.so/dev-tools/agents/overview

What you can do today

Since your agents are running on OpenClaw, you could build a thin integration layer using the Plane Agent SDK (Node.js / Python) that bridges your OpenClaw agents with Plane. The integration would:

1. Receive Plane webhooks when work items are assigned to or mention your agent
2. Forward the task context to your OpenClaw agent
3. Stream progress back to Plane as thoughts/actions/responses

This gives you assignment-based triggers, execution tracking, and conversation threads — all within the Plane UI — with minimal glue code.

What's coming next

We're exploring a no-code/low-code agent management layer that would make it easier to connect external agent platforms like OpenClaw to Plane without writing integration code. This would complement the existing SDK-based approach and move closer to the "Mission Control" experience you're describing.