A high-scale processing engine designed to transform massive dumps of mixed content into executed tasks and structured reports using parallel sub-agents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install batch-executor
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-executor 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 Executor?
Batch Executor is a specialized framework within the Openclaw Skills ecosystem built for corpus-scale task execution. Unlike simple extractors, it is designed to handle thousands of items—from Google Drive dumps and ChatGPT history to years of Apple Notes—ensuring no data is lost during ingestion. It serves as the primary execution engine for high-volume workloads where manual oversight of every single item is impossible.
The skill operates on a robust architecture that prioritizes data integrity and systematic progress. By using a multi-phase approach of classification, ICE (Impact, Cost, Exploit) scoring, and parallel execution, Batch Executor transforms raw noise into a structured value stack. It manages the lifecycle of AI sub-agents, ensuring that heavy tasks are offloaded efficiently while maintaining a clear audit trail via checkpoints and git commits.
Batch Executor Use Cases
- Processing Google Drive folder dumps containing mixed documents, notes, and spreadsheets.
- Analyzing and executing tasks from massive ChatGPT conversation exports (3K+ prompts).
- Migrating and categorizing years of ideas from Apple Notes dumps.
- Managing any input exceeding 20 items or 10,000 tokens that requires systematic execution rather than just simple analysis.
How Batch Executor Works
- Ingest: Saves all raw input to a local directory and creates a manifest to ensure no data is lost before processing begins.
- Classify: Categorizes each item (Task, Idea, Reference, Decision, etc.) and estimates the required effort from trivial to heavy.
- Triage: Scores items using the ICE framework to prioritize high-impact, low-cost actions and creates a wave-based execution plan.
- Execute: Spawns up to three concurrent sub-agents for heavy tasks while executing quick tasks inline to maximize throughput.
- Reconcile: Verifies completions, retries failed tasks with different approaches, and cleans up stale agent processes.
- Report: Generates a comprehensive final report detailing the value stack, discovered patterns, and immediate action items.
Batch Executor Setup
To initialize the environment for this Openclaw Skills component, ensure your directory structure is prepared for corpus storage:
mkdir -p systems/batch-executor/corpus
mkdir -p systems/batch-executor/reports
Once the directories are set, you can trigger the executor by pointing it to a source file or folder dump. The skill will automatically begin the INGEST phase and generate a manifest.md for tracking.
Batch Executor Data Schema & Taxonomy
Batch Executor maintains a strict directory and metadata taxonomy to ensure resume-ability and transparency:
| File Type | Path | Purpose |
|---|---|---|
| Raw Corpus | systems/batch-executor/corpus/YYYY-MM-DD-SOURCE.md |
Persistent storage of original input data. |
| Manifest | systems/batch-executor/corpus/manifest.md |
Tracking table for item status, type, and ICE scores. |
| Execution Plan | Internal Markdown | Defines the waves of parallel sub-agent tasks and dependencies. |
| Final Report | systems/batch-executor/reports/YYYY-MM-DD-SOURCE-report.md |
Summary of outcomes, patterns, and parked items. |
| Value Stack | systems/batch-cognition/value-stack.md |
Shared file for high-impact insights across skills. |
Batch Executor Advanced Features
- Parallel Sub-Agent Management: Orchestrates up to 3 concurrent agents with automatic context injection from cross-referenced corpus items.
- First-Class Resume Support: If a session is interrupted, the skill re-reads the manifest to continue exactly where it left off.
- Automated Checkpoints: Every 5 completed items, the skill performs a git commit and provides a status update to the user.
- ICE Scoring Logic: Fast-track triage system that calculates priority based on Impact × Cost × Exploit metrics.
- Cross-Skill Integration: Automatically routes finished results to the recorder skill and updates shared cognition value stacks.
SKILL.md
Loading
Related Openclaw Skills
Financial Statement Review Skill
A professional AI-driven financial statement review tool for comprehensive compliance auditing and tax risk identification based on Chinese accounting standards.
Spawner
Spawner is a high-performance orchestration skill that automates the hiring, configuration, and monitoring of AI sub-agents to execute complex job board tasks.
Orchestrator v2: Strategic Project Manager
A strategic management agent that transforms vague user requests into comprehensive, multi-wave execution plans by inferring hidden requirements.
memory-bridge
A persistence layer that enables AI agents to remember user preferences, project context, and task progress across session restarts.
Task Extractor
The Task Extractor is a robust workflow skill that captures, organizes, and verifies multiple instructions from a single prompt to ensure zero task drop.
Kawaii Art Generator
An AI-powered generator for creating cute characters, chibi illustrations, and Japanese kawaii aesthetics.
Featured*
