A powerful workflow skill that transforms raw requirements into structured, dependency-ordered implementation plans and orchestrates parallel execution via subagents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install batch-plan-execute
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 batch-plan-execute 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 Batch Plan Execute?
Batch Plan Execute is a sophisticated capability designed for modern AI agents within the Openclaw Skills ecosystem. It solves the challenge of translating ambiguous requirement documents or chat instructions into actionable, modular code changes. By separating the logic into distinct plan and execute modes, it ensures that developers have full oversight before any code is modified. The skill identifies implementation boundaries, models strict dependency graphs, and manages a hidden state file to track progress across complex workstreams.
This skill is particularly valuable for large-scale refactors or feature additions where multiple files and modules are affected. It leverages subagents for both analysis and implementation, allowing for high-concurrency workflows without sacrificing architectural integrity. By using Openclaw Skills like this, teams can ensure that AI-driven development remains organized, traceable, and strictly aligned with project requirements.
Batch Plan Execute Use Cases
- Scaling high-level feature requests into granular, multi-module implementation strategies.
- Managing complex refactoring tasks that require a specific sequence of foundation and feature updates.
- Automating the revision of existing plans based on peer review feedback or updated requirement documents.
- Orchestrating parallel code implementation across independent modules to accelerate development cycles.
How Batch Plan Execute Works
- Input Resolution: The skill identifies the requirement source from chat text, markdown files, or existing plan directories.
- Module Extraction: It analyzes requirements to define implementation units based on functional ownership rather than just file structure.
- Dependency Modeling: Modules are classified as foundation, feature, or follow-on to build a non-cyclic dependency graph.
- Planning Phase: Subagents generate detailed markdown plans for each module, including affected code areas and risks.
- Execution Gating: The system pauses for an explicit implementation command, ensuring a human-in-the-loop review process.
- Orchestrated Implementation: Implementation subagents are dispatched to modify code in parallel layers according to the dependency graph.
- Verification and Integration: The main agent runs validation checks to ensure all modules integrate correctly and meet the original requirements.
Batch Plan Execute Setup
To integrate this skill into your environment, ensure your workspace is prepared for structured plan storage. This skill is a core part of the Openclaw Skills suite.
# Create the standard plans directory
mkdir -p plans/
# Ensure your agent has permissions to spawn subagents
# The skill will automatically generate .batch-plan-state.json upon first use
Batch Plan Execute Data Schema & Taxonomy
The skill maintains a rigorous data structure within the plans/ directory to ensure consistency and traceability.
| File/Element | Type | Description |
|---|---|---|
<module-slug>.md |
Markdown | The primary implementation plan for a specific module. |
<module-slug>.rev-<n>.md |
Markdown | Versioned revision files reflecting requirement changes or review notes. |
.batch-plan-state.json |
JSON | The state machine tracking requirement hashes, module statuses, and action lineage. |
requirement_hash |
Metadata | Used to detect when the source document has changed and requires a plan revision. |
status |
Enum | Tracks modules as active or obsolete within the current implementation cycle. |
Batch Plan Execute Advanced Features
- Intelligent Revision Detection: Automatically detects HTML comments as review notes and incorporates them into plan revisions.
- Dependency-Aware Parallelism: Maximizes throughput by running subagents in parallel only within the same dependency layer.
- Automated Rename Matching: Tracks module lineage even if filenames change, using body hash analysis.
- Failure Isolation: Stops execution immediately if dependency cycles or ambiguous ownership are detected, preventing broken builds.
- State Reconstruction: Capable of bootstrapping the current progress from existing artifacts if the state file is missing or corrupted.
SKILL.md
Loading
Related Openclaw Skills
Nano Banana + Veo
An advanced AI workflow that generates premium static images using Gemini 3 Pro and transforms them into cinematic videos via the Veo 3.1 engine.
Work Todo
A professional task management system for AI agents that handles task lifecycle tracking, project mapping, and automated performance reporting.
RF Engineering Assistant
A comprehensive technical reference and calculation engine for radio frequency design, wireless propagation modeling, and hardware certification.
Signal Processing AI Expert
A professional-grade knowledge base and toolset for signal processing, covering everything from basic LTI systems to advanced MIMO and OFDM communication engineering.
AI Interview Assistant
An intelligent multi-stage interview agent that manages technical profiles, JD matching, and structured 5-question mock interview cycles with real-time scoring.
Embedded Dev
A comprehensive full-stack embedded development skill covering everything from low-level hardware registers to high-level RTOS application logic.
Featured*
