A multi-project Kanban framework designed for AI agents to maintain persistent task states and deterministic board workflows.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install kanban
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 kanban 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 Kanban?
The Kanban skill enables AI agents to build and maintain sophisticated task management systems with persistent routing across conversations. By integrating this into your environment via Openclaw Skills, you ensure that the agent can autonomously discover project boards, follow strict processing rules, and maintain an audit log of every change. It bridges the gap between ephemeral chat sessions and long-term project tracking by storing metadata in a structured, local hierarchy.
This system allows for both centralized management in a home directory and project-local storage, providing flexibility for different development environments. The architecture is built around a discovery protocol that ensures any agent session can pick up exactly where the last one left off without losing context or task history.
Kanban Use Cases
- Tracking progress across multiple repositories with a centralized discovery index.
- Maintaining task continuity for AI agents across different chat sessions or IDE restarts.
- Enforcing strict project management rules and work-in-progress limits within an automated workflow.
- Creating an audit trail for all task movements and status updates to monitor project velocity.
How Kanban Works
- Discovery: The agent runs a discovery protocol to identify the relevant project board from the index or local workspace.
- Context Resolution: Current board state and project-specific rules are loaded to ensure processing alignment.
- Deterministic Processing: Tasks are moved or updated following a fixed decision order to handle dependencies and priorities accurately.
- Atomic Persistence: Changes are written to the board file and recorded in the project log simultaneously to prevent data corruption.
- Routing Update: The global registry is updated with the last used timestamp and path to preserve session continuity for future interactions.
Kanban Setup
To begin using the Kanban skill, ensure your directory structure is ready. The agent will initialize the necessary templates upon first run once the directory is detected.
# Initialize the core kanban directory
mkdir -p ~/kanban/projects
# The agent will silently read setup.md and request confirmation
# to populate templates and memory files during the first session.
Kanban Data Schema & Taxonomy
The skill utilizes a structured file system to ensure data integrity and easy parsing by Openclaw Skills. The data is organized as follows:
| Component | Path | Function |
|---|---|---|
| Registry | ~/kanban/index.md |
Maps project IDs to their physical board locations and aliases. |
| Board | project-id/board.md |
Markdown file containing the active task cards and their current states. |
| Rules | project-id/rules.md |
Defines the specific lanes, priorities, and movement policies for a project. |
| Log | project-id/log.md |
Persistent record of all card transitions and edits for auditing. |
Kanban Advanced Features
- Multi-agent continuity through persistent routing memory stored in the global index file.
- Support for custom project-local modes using a
.kanban/folder within specific workspaces. - Deterministic conflict resolution that prevents the agent from skipping blockers or violating WIP limits.
- Seamless integration with other Openclaw Skills like daily-planner and workflow for end-to-end automation.
SKILL.md
Loading
Related Openclaw Skills
JWT Authentication Security
A comprehensive technical skill for implementing and managing secure JSON Web Token (JWT) authentication workflows.
Job Search
An intelligent agent skill for managing the end-to-end job application lifecycle through research, material customization, and pipeline tracking.
JEE Exam Prep Assistant
A specialized AI assistant designed to optimize preparation for India's Joint Entrance Examination (JEE) through data-driven tracking and strategy.
Japan
A specialized AI skill providing hyper-local travel intelligence, itinerary planning, and logistical expertise for trips across Japan.
Keras
A comprehensive deep learning skill for building, training, and troubleshooting neural networks using Keras and TensorFlow patterns.
Linux Kernel Guardrails
A specialized technical resource for preventing common Linux kernel development pitfalls like context violations and memory errors.
Featured*
