Plans Methodology for Openclaw

A standardized framework for managing the lifecycle of complex tasks and multi-agent coordination within an AI workspace.

cramtek
v1.0.0
Feb 21, 2026
0
1.3k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install plans-methodology

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install plans-methodology 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 Plans Methodology?

The Plans Methodology is a robust system designed for AI coding agents and teams to track significant work through a formal lifecycle. By utilizing a standardized directory structure and document templates, it ensures that every project—from initial proposal to final architectural design—is documented, approved, and executed with clarity.

This system is a core component of high-performing Openclaw Skills setups, providing the necessary hierarchy and traceability for complex software development and research tasks. It bridges the gap between high-level intent and low-level task execution by enforcing a consistent documentation standard across the entire agent workspace.

Plans Methodology Use Cases

  • Coordinating complex features across multiple specialized AI sub-agents in a team environment.
  • Managing multi-phase projects that require formal stakeholder or lead-agent approval before execution.
  • Creating a historical audit trail of architectural decisions, risks, and task completion for long-term projects.
  • Tracking progress during long-running sessions or facilitating seamless handovers between different agent roles.

How Plans Methodology Works

  1. Initialize a plan by creating a new dated folder within the draft directory of the plans workspace.
  2. Draft the core documentation including the problem statement in proposal.md, technical architecture in design.md, and actionable items in tasks.md.
  3. Submit the proposal for approval based on your established hierarchy, moving the plan to the approved directory once cleared for execution.
  4. Execute tasks by moving the folder to the in_progress directory and updating the task list with completion dates and notes.
  5. Finalize the work by moving the entire folder to completed or archived for future reference and knowledge retrieval.

Plans Methodology Setup

To implement this methodology within your workspace, create the base directory structure using the following commands:

mkdir -p plans/{draft,approved,in_progress,completed,archived}
touch plans/README.md

Once the structure is in place, define your approval authority in the plans/README.md to ensure the agent knows who has the power to move plans from draft to approved status. This setup is essential for managing sophisticated Openclaw Skills workflows.

Plans Methodology Data Schema & Taxonomy

The methodology organizes data into a hierarchical file structure for maximum readability by both humans and AI agents:

File/Folder Purpose
plans/ The root directory for all project tracking and methodology state.
YYYY-MM-DD-name/ A unique container for a specific project or initiative.
specs/ Contains raw requirements, problem statements, and user stories.
proposal.md Documents the 'Why' (justification, impact, risks, and approval status).
design.md Documents the 'How' (architecture, flows, constraints, and dependencies).
tasks.md Documents the 'What' (checklists, owners, and delegation links).

Plans Methodology Advanced Features

  • Cross-agent delegation using bidirectional parent-child plan referencing in proposal and task files.
  • Support for hierarchical approval chains, allowing for strategic oversight by a CEO agent or human operator.
  • Contextual session recovery by scanning the in_progress directory upon agent startup to resume work immediately.
  • Multi-agent coordination protocols through shared file-based state management and standardized folder movements.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*