A high-integrity three-layer orchestrator that separates task dispatching from execution and verification to ensure reliable project completion.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install dispatcher-v2
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install dispatcher-v2 using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is dispatcher-v2?
dispatcher-v2 is the primary orchestrator for the OpenClaw architecture, designed to solve the common issue of AI agents claiming completion without actual execution. By enforcing a strict separation of concerns, the dispatcher acts solely as a manager that delegates tasks to specialized sub-agents while mandating independent verification.
This skill is a cornerstone of the Openclaw Skills ecosystem, providing a robust framework where execution, verification, and status tracking are handled by distinct, specialized layers. It ensures that no task is reported as finished until a separate verifier agent confirms the work against the intended goals.
dispatcher-v2 Use Cases
- Managing complex multi-file development tasks across different specialized agents.
- Ensuring core configuration changes in files like RULES.md or AGENTS.md are verified before reporting success.
- Recovering from system crashes using persistent progress tracking files that act as a source of truth.
- Synchronizing project state between external IDEs like Cursor and the agent environment.
How dispatcher-v2 Works
- The dispatcher receives a task and categorizes it into a difficulty level (S, A, B, C, or D) based on complexity.
- A detailed plan and execution path are presented to the user for confirmation and authorization.
- A progress file is initialized in the memory directory to serve as the persistent source of truth throughout the lifecycle.
- The dispatcher spawns a specialized executor sub-agent to perform the actual coding or configuration work.
- Upon completion of execution, a verifier sub-agent is spawned to independently audit the results against the progress checklist.
- If the work is PASSED, the dispatcher performs a final sync to daily memory and synchronization files before notifying the user.
dispatcher-v2 Setup
To integrate this skill into your environment, ensure your workspace has the required directory structure for memory persistence. This skill requires the Openclaw Skills runtime to handle sub-agent spawning.
# Create the required memory directory
mkdir -p ~/.openclaw/workspace/memory
# Initialize the progress and regression tracking files
touch ~/.openclaw/workspace/regressions.md
touch ~/.openclaw/workspace/CURSOR_SYNC.md
dispatcher-v2 Data Schema & Taxonomy
The skill organizes its operations through a strict file-based taxonomy to ensure persistence and reliability:
| Component | Path | Purpose |
|---|---|---|
| Progress File | memory/progress-<slug>.md |
Stores task goals, start times, and current execution status. |
| Daily Log | memory/YYYY-MM-DD.md |
Appends a record of completed activities for long-term memory. |
| Sync File | CURSOR_SYNC.md |
Manages state synchronization between OpenClaw and the IDE. |
| Regression Log | regressions.md |
Records protocol violations and verification failures for future optimization. |
dispatcher-v2 Advanced Features
- Automated Crash Recovery: Active progress monitoring allows the dispatcher to resume tasks after session interruptions or model timeouts.
- Hierarchical Task Grading: Intelligent categorization ensures that simple queries bypass heavy protocols while complex changes require strict verification.
- Multi-Agent Spawning: Dynamically assigns work to specialized agents (alpha, bravo, charlie, etc.) based on the technical domain of the task.
- Fail-Safe Mechanisms: Includes retry logic (up to 3 times) for failed verifications and blocks tasks requiring human intervention if automated fixes fail within Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
X Search Highlights
A sophisticated extraction tool that identifies and ranks high-value posts from X based on engagement metrics like retweets and likes.
Plume AI Notecard Service
A high-performance AI service designed to convert topics, articles, and reference images into structured, visually optimized information cards and infographics.
Monday.com Integration
A comprehensive GraphQL-based integration for Monday.com that provides AI agents with managed OAuth and full control over project management workflows.
Construction Mediation Knowledge Graph
A professional AI-driven knowledge graph designed to provide systematic support for the entire lifecycle of construction commercial mediation and dispute resolution.
OpenClaw Mastery & Skills Guide
A comprehensive framework for deploying and extending OpenClaw, a local-first, model-agnostic AI assistant platform built for privacy and autonomy.
Safe Evolver
Safe Evolver is a robust AI evolution engine that analyzes execution logs to identify performance bottlenecks and apply safety-constrained behavioral improvements.
Featured*
