PLAN() Planning Decomposition Skill for Openclaw

PLAN() is an Openclaw Skills planning agent that transforms vague goals into structured, executable task plans with dependencies, timing, resources, and completion criteria.

zaynpeng
v1.0.0
Jul 28, 2026
0
306
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install zayn-plan

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 zayn-plan 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 PLAN() Planning Decomposition Skill?

PLAN() is a planning decomposition skill built to convert ambiguous user goals into a clear, actionable execution plan. It forces the input into a parameter state table, validates evidence, and only then produces a structured output that includes goals, key paths, staged tasks, dependencies, resources, constraints, and completion standards.

This Openclaw Skills workflow is designed for situations where a user has core material and wants a reliable decision or action path, not a brainstorm or wishlist. By enforcing minimum operating conditions, explicit boundary checks, and human judgment when information is incomplete, PLAN() helps keep plans realistic, verifiable, and execution-ready.

PLAN() Planning Decomposition Skill Use Cases

  • Breaking down a vague objective into a step-by-step execution plan
  • Turning user-provided evidence into structured task sequencing
  • Identifying dependencies, priorities, risks, and resource limits before action
  • Producing a plan only when minimum input conditions are met
  • Preparing decision support when the user needs a clear path forward
  • Preventing over-scoped, low-priority, or assumption-heavy planning in Openclaw Skills

How PLAN() Planning Decomposition Skill Works

  1. Parse the user request and map the input into a parameter state table.
  2. Check whether required fields are present: target, current state, cutoff time, and scope.
  3. Validate optional fields such as resources, constraints, priority, dependencies, owner, and completion criteria.
  4. Confirm evidence priority and ensure at least one reliable source is available.
  5. Apply judgment rules: keep goals specific, tasks executable, dependencies explicit, and task count controlled.
  6. Run risk checks to avoid wishlist-style plans, overloaded low-priority tasks, ignored dependencies, and assumed resource sufficiency.
  7. If required information is missing, request clarification instead of fabricating a full plan.
  8. When minimum conditions are satisfied, output the final structure: completeness conclusion, parameter table, goal definition, key paths, staged tasks, dependency relationships, resources and limits, completion criteria, and next steps.

PLAN() Planning Decomposition Skill Setup

PLAN() is a skill-based workflow rather than a package with a traditional install command. To use it effectively in Openclaw Skills, follow this configuration flow:

  1. Provide the core objective, current state, deadline, and scope.
  2. Include at least one reliable piece of evidence or source material.
  3. Add optional context where available: resources, constraints, priority, dependencies, responsible person, and completion criteria.
  4. Avoid asking the skill to infer missing facts as final truth; instead, supply the missing parameters when prompted.
  5. Use the skill only for planning and decomposition, not for unrelated tasks outside its boundary.
# Example input preparation for PLAN()
cat > plan-input.txt << 'EOF'
Target: ...
Current state: ...
Deadline: ...
Scope: ...
Evidence: ...
Optional context: resources, constraints, priority, dependencies, owner, completion criteria
EOF
# Then pass the structured request into your Openclaw Skills workflow
openclaw skill run zayn-plan --input plan-input.txt

PLAN() Planning Decomposition Skill Data Schema & Taxonomy

PLAN() organizes information into a parameter-first planning schema.

Field group Required fields Optional fields Purpose
Core parameters Target, current state, deadline, scope — Establishes the planning boundary
Context parameters — Resources, constraints, priority, dependencies, owner, completion criteria Adds execution realism and accountability
Evidence layer At least one reliable source Additional supporting materials Prevents assumption-driven planning
Validation state Fully matched, partially matched, missing, conflicting, pending verification — Tracks input completeness before formal analysis

Output taxonomy

  • Parameter completeness conclusion
  • Parameter state table
  • Goal definition
  • Key paths / critical routes
  • Stage-by-stage tasks
  • Dependency relationships
  • Resources and constraints
  • Completion criteria
  • Next step recommendation

State rules

  • Missing parameters must be explicitly requested, not auto-filled.
  • Conflicting parameters must be preserved with source notes and marked for verification.
  • Pending information cannot be promoted to confirmed facts.
  • Formal analysis begins only after minimum operating conditions are met.

PLAN() Planning Decomposition Skill Advanced Features

  • Parameter state table enforcement before analysis begins
  • Evidence-first validation to reduce assumption drift in Openclaw Skills
  • Minimum operating conditions gate for safer planning outputs
  • Conflict preservation logic that avoids premature resolution
  • Human judgment boundary that prevents automatic scope expansion
  • Executable task decomposition with dependency awareness
  • Risk controls against wishlist-style planning and overloaded task stacks
  • Clear final output structure optimized for downstream execution

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*