Workspace

Agent Fleet

The sovereign control plane for your OpenClaw agent fleet. Agents, projects, tasks, skills, and providers as connected entities — not daemon logs.

Record types

The structured data this adds to your pod.

  • Agent

    A deployed OpenClaw agent — your running AI entity

    • Status
    • Health
    • Runtime
    • Version
    • Primary Model
    • Provider
    • Owner
    • Primary Channel
    • +7
  • Project

    A bounded body of work delegated to one or more agents

    • Status
    • Priority
    • Assigned Agent
    • Human Owner
    • Start Date
    • Deadline
    • Health
    • Type
    • +3
  • Task

    An atomic work item assigned to an agent

    • Status
    • Priority
    • Assigned Agent
    • Project
    • Type
    • Due Date
    • Effort Estimate (hrs)
    • Token Estimate (K)
    • +4
  • Skill

    An installed agent capability — a ClawHub skill, custom script, or API integration

    • Trust Level
    • Category
    • Source URL
    • Version
    • Install Date
    • Last Audit Date
    • VirusTotal Status
    • Maintainer
    • +2
  • Provider

    An LLM or API provider — tracks cost, key health, and agent usage

    • Type
    • Key Health
    • Model / Endpoint
    • Cost per 1K Input Tokens ($)
    • Cost per 1K Output Tokens ($)
    • Daily Budget Ceiling ($)
    • Actual Spend This Week ($)
    • Base URL / Endpoint
    • +1
  • Person

    Team member or agent owner

    • Role
    • Email
    • Telegram Handle
    • Current Focus

Views

  • Agent FleetBoard
  • Agent → Projects → TasksGraph
  • Task BoardBoard
  • Project Command CenterTable
  • Skill RegistryTable
  • Provider DashboardTable
  • Relationship GraphGraph
  • Agent TimelineTimeline
  • Cost ViewTable
  • Untrusted Skills AlertTable
  • BacklogTable
  • Agent Health MonitorBoard
  • Session LogsTable
  • Provider Key HealthBoard
  • My Active WorkTable
  • AgentDashboard
  • TaskDashboard
  • ProjectDashboard
  • SkillDashboard
  • ProviderDashboard

Playbooks

Repeatable runs your agent can start for this project.

  • Agent Health Check
  • Security Audit

First conversation

After install, your agent sets the project up with you. It starts here:

Capture the agent fleet the user runs — the agents themselves, the model providers behind them, the skills they carry, and the projects they work — so the workspace reflects their real AI operation.

  • “Which agents are you running right now, and what's each one for?”
  • “What models and providers power them, and are there budget limits?”
  • “What skills do your agents carry, and which projects are they pointed at?”