run

Launch drover error monitoring with pre-flight checks and parallel triage.

Updated Feb 28, 2026

Search Tags:#agent orchestration #error monitoring #state checkpointing #drover #triage automation #kanban verification #dashboard startup

Ready to run

npx skills add https://github.com/cosmicdreams/claude-plugins --skill run-cosmicdreams

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: run
Source: https://github.com/cosmicdreams/claude-plugins/tree/main/drover/skills/run
Command: npx skills add https://github.com/cosmicdreams/claude-plugins --skill run-cosmicdreams

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It removes the friction of manually validating setup, starting the drover dashboard, and running a full triage cycle across environments so you can start monitoring errors immediately.

Core Features & Use Cases

Bootstrap full drover pipeline: Performs environment/config validation, launches the dashboard UI, and runs a complete first triage cycle across configured environments.
Parallel agent triage with team orchestration: Creates a triage team, spawns one triage agent per environment, collects results, and records a consolidated checkpoint.
Post-triage verification and state tracking: Verifies whether prior fingerprints reappeared and updates the state store for future monitoring.

Quick Start

Start full drover monitoring by telling the assistant: "run drover".

Frequently Asked Questions about run

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

FAQPage Schema

How do I start drover error monitoring end-to-end?▼

To start drover error monitoring, you trigger a full run that validates configuration and environment health, launches the dashboard UI, and executes a complete triage cycle across all configured environments. You can initiate this by simply telling the assistant to run drover.

How does parallel agent triage work for error monitoring?▼

Parallel agent triage works by orchestrating a team that spawns one triage agent per configured environment. The agents collect results concurrently, and the system records a consolidated checkpoint to update the state store for future monitoring and verification.

What is drover state checkpointing and verification?▼

Drover state checkpointing is the process of updating the ~/.claude/drover.state.jsonl file after a triage cycle. Verification checks whether prior error fingerprints reappeared across environments, ensuring accurate state tracking and actionable next steps.

Do I need to run health checks before bootstrapping the drover dashboard?▼

Yes, pre-flight checks are enforced before bootstrapping the drover dashboard. The system performs environment validation and runs DDEV health verification to ensure your setup is healthy before starting the UI and executing the initial error triage.

Can I use drover for first-time error triage setup across multiple environments?▼

Yes, drover is applicable to first-time setup and initial error triage across multiple configured environments. It coordinates TeamCreate and TeamDelete parallel triage agents to collect results, update the state store, and present actionable outcomes.