An AI-driven strategic planning consultant that applies Recursive Agents and Landmarks Strategic-Tactical Planning (RALSTP) to decompose complex workflows and identify agent dependencies.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install ralstp-consultant
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 ralstp-consultant 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 RALSTP Consultant?
This skill implements the Recursive Agents and Landmarks Strategic-Tactical Planning framework based on the 2024 PhD thesis by Dorian Buksz from King's College London. It allows developers and project managers to analyze complex systems by identifying active agents, passive objects, and the entanglement between resources. By utilizing this within Openclaw Skills, users can transform abstract problem descriptions into actionable strategic and tactical plans.
Whether you are managing a fleet of autonomous vehicles or planning a large-scale cloud migration, this consultant provides the mathematical rigor needed to calculate complexity and identify critical path landmarks. It bridges the gap between high-level strategic goals and low-level tactical execution, ensuring that resource contention and agent dependencies are addressed before implementation begins.
RALSTP Consultant Use Cases
- Planning large-scale cloud or VM migrations with minimal downtime.
- Designing multi-agent robotic or software systems with shared resource constraints.
- Orchestrating complex marketing launches involving multiple departments and dependencies.
- Analyzing PDDL (Planning Domain Definition Language) files for automated planning research.
- Decomposing monolithic project workflows into parallelizable tasks.
How RALSTP Consultant Works
- The user provides a problem description or PDDL domain files to the Openclaw Skills environment.
- The consultant identifies dynamic agents by analyzing action effects and distinguishes them from passive objects.
- A dependency graph is generated to map how agents satisfy preconditions for one another.
- The skill calculates entanglement scores based on shared predicates and resource conflicts.
- It identifies fact and action landmarks to establish the mandatory sequence of events.
- A two-phase plan is produced: a Strategic Phase for abstract goals and a Tactical Phase for detailed execution.
RALSTP Consultant Setup
To get started with this skill in your environment, follow these steps:
# Clone the repository containing the RALSTP skill
git clone https://github.com/openclaw/ralstp-consultant.git
cd ralstp-consultant
# Install necessary dependencies for Formal Mode
pip install -r requirements.txt
Once installed, you can invoke the consultant directly through your Openclaw Skills interface by describing a workflow or providing PDDL paths.
RALSTP Consultant Data Schema & Taxonomy
The skill organizes analysis into the following structure:
| Component | Description |
|---|---|
| Agents | List of objects with dynamic types and their roles. |
| Passive Objects | Static entities acted upon by agents. |
| Dependency Graph | Mapping of Independent, Dependent, and Conflicting relationships. |
| Metrics | Complexity scores based on Agent Count and Entanglement Factor. |
| Landmarks | Required fact and action sequences (The Critical Path). |
RALSTP Consultant Advanced Features
- Dual-mode analysis: Supports both Natural Language processing and formal PDDL mathematical extraction for users of Openclaw Skills.
- Complexity Scoring: Implements the Buksz Complexity Score (Agent Count x Entanglement Factor) for rigorous risk assessment.
- Decomposition Logic: Suggests optimal ways to split tasks by agent type or landmark for maximum parallelization.
- Resource Contention Detection: Automatically flags potential deadlocks where multiple agents require the same location or state simultaneously.
SKILL.md
Loading
Related Openclaw Skills
Quality Gateways (Generic Web + API Applications)
A language-agnostic framework for enforcing six universal quality gateways across backend APIs and web frontends using LLM-friendly operational standards.
Top 80 MSBuild Commands for .NET & ASP.NET
A prioritized collection of 80 essential MSBuild command templates for .NET and ASP.NET CLI workflows.
MetriLLM
MetriLLM is a powerful CLI benchmarking tool designed to evaluate local LLMs for speed, response quality, and hardware compatibility.
tex-render
A utility for AI agents to render LaTeX mathematical notation into high-quality, viewable images.
Echo Repeater
A minimal utility skill designed to echo user input back for testing skill routing and message integrity.
Glin Profanity MCP Server
A Model Context Protocol server providing AI assistants with robust tools for profanity detection, text censoring, and content safety analysis.
METADATA
Featured*
