task-dashboard

Inspect swarm runtime state and summarize task health into a concise dashboard.

3|2|Updated Mar 6, 2026
One-click install
npx skills add https://github.com/cloga/optimus-code --skill task-dashboard-cloga
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: task-dashboard
Source: https://github.com/cloga/optimus-code/tree/main/optimus-plugin/skills/task-dashboard
Command: npx skills add https://github.com/cloga/optimus-code --skill task-dashboard-cloga

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Swarm observability is fragmented across task manifests, agent statuses, and lock files. This Skill provides a single, current view by turning those signals into a concise dashboard that highlights health and progress.

Core Features & Use Cases

  • Inspect and summarize swarm runtime state from .optimus/state/task-manifest.json, agent frontmatters in .optimus/agents/, and lock files to surface running, completed, verified, partial, and failed tasks.
  • Flag stale tasks and provide recent completion history to aid troubleshooting and auditing.
  • Generate a compact overview suitable for user requests or during pre-delegation checks to avoid duplicate work.

Quick Start

Ask the Master Agent to generate a concise swarm dashboard from the current state in .optimus/state and .optimus/agents.

Frequently Asked Questions about task-dashboard

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get a dashboard of swarm runtime state for ongoing tasks?

To get a dashboard of swarm runtime state, this Skill inspects `.optimus/state/task-manifest.json`, agent frontmatters, and lock files to summarize running, completed, verified, partial, and failed tasks into a concise overview.

Why does my task manifest show a task as running when it seems stuck?

If a task manifest shows a task as running when it seems stuck, this Skill reads agent frontmatters and lock files to flag stale tasks and identify completed or failed work, helping you troubleshoot async dispatches that are no longer active.

What is the best way to avoid duplicate work before delegating new tasks to agents?

The best way to avoid duplicate work before delegating new tasks is to generate a pre-delegation dashboard that checks current swarm status from `.optimus/state` to verify running, completed, and verified tasks.

Can I check the health of async dispatched tasks from .optimus state files?

Yes, you can check the health of async dispatched tasks by reading `.optimus/state/t3-usage-log.json` and task manifests to surface recent completions, flag stale tasks, and report current swarm status.

Does this dashboard tool require specific dependencies to aggregate agent statuses?

No specific dependencies are required to aggregate agent statuses. The tool directly reads existing state files, agent frontmatters from `.optimus/agents/*.md`, and lock files to determine task health without external packages.