Planning Mode is a sophisticated project architectural framework that transforms vague ideas into structured execution plans through dynamic brainstorming and autonomous research.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install situated-planning-mode
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 situated-planning-mode 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 Planning Mode?
Planning Mode serves as a high-level strategic layer for AI agents, functioning as a collaborative brainstorming partner. It is built on the philosophy that commands and descriptions are inseparable; instructions without context lead to execution errors, while context without instructions leads to inaction. By integrating this skill into Openclaw Skills environments, developers can ensure that every project phase—from initial discovery to final validation—is documented with clear consequences, risks, and resource costs.
The skill excels at bridging the gap between a user's vision and a machine's execution requirements. It prevents premature coding by enforcing a staged workflow where knowledge gaps are identified and filled. If the agent encounters a technical unknown, it autonomously triggers subagent research to provide the user with data-driven options, ensuring the final plan is both feasible and comprehensive.
Planning Mode Use Cases
- Initial project discovery and scope definition for new software applications.
- Technical feasibility analysis for complex AI integrations.
- Strategic brainstorming sessions to identify project risks, costs, and scalability.
- Transforming high-level user requests into detailed developer roadmaps.
How Planning Mode Works
- The skill is triggered when a user proposes a new project or a complex task requiring structure.
- The agent analyzes the project context to determine the project type and current development stage.
- A dynamic question tree is generated, spanning six fixed stages: Discovery, Analysis, Design, Review, Develop, and Validate.
- During each stage, the agent presents options that include both the action (command) and its context (description).
- If information is missing, the agent uses sessions_spawn to launch a subagent for autonomous research.
- Once all stages are completed, the agent generates a final JSON summary containing the full context and execution commands.
Planning Mode Setup
To activate this skill within your environment, ensure the following configuration is present in your skill directory. This skill relies on the sessions_spawn and memory tools typical of Openclaw Skills implementations.
# Ensure your agent has access to these required tools:
# sessions_spawn (for subagent research)
# memory_search (for context retrieval)
# memory_get (for specific data fetching)
Planning Mode Data Schema & Taxonomy
The skill produces a structured JSON output to transition from planning to execution. The schema tracks the progression through the six mandatory stages.
| Key | Description |
|---|---|
| action | Indicates if the planning is 'planning_completed' or 'planning_incomplete' |
| result | Success or failure status of the planning session |
| stages | A mapping of the 6 stages (Discovery to Validate) and their specific selections |
| summary | A comprehensive markdown-formatted description of the final plan |
| next_action | The logical next step for the execution agent |
Planning Mode Advanced Features
- Autonomous subagent research to fill knowledge gaps without user intervention.
- Multi-dimensional analysis covering risks, costs, time-to-launch, and scalability.
- Dynamic question generation that evolves based on previous user responses.
- Cross-session memory integration to maintain consistency across different Openclaw Skills project plans.
SKILL.md
Loading
Related Openclaw Skills
pdfly CLI
A robust Python-powered command-line interface for merging, splitting, compressing, and signing PDF documents.
Micro Cluster Momentum Trader
A sophisticated trading strategy that detects cross-coin directional convergence in 5-minute Polymarket bundles to execute micro-sized momentum continuation trades.
Micro Pulse Revert Trader
A specialized trading skill that detects extreme retail sentiment pulses on Polymarket's 5-minute crypto markets to execute automated mean-reversion trades.
Polymarket Micro Coin Lag Trader
A specialized trading skill that exploits price discovery delays between Bitcoin and altcoins in Polymarket 5-minute interval markets.
Git Reporter
An intelligent git activity analyzer that transforms raw commit logs into professional standups, weekly reports, and sprint retrospectives.
Weekly Report Genius
An AI-powered professional writing assistant that converts casual work notes into high-impact corporate weekly reports and summaries.
Featured*
