🌴 Simple utility for managing parallel Claude Code instances
git clone https://github.com/jlowin/claude-wt.gitThe claude-wt skill is a powerful automation tool designed for managing multiple instances of Claude Code in parallel. This utility simplifies the process of orchestrating various AI tasks simultaneously, allowing developers and AI practitioners to enhance their workflow efficiency. By enabling the parallel execution of Claude Code instances, users can optimize resource allocation and improve overall productivity in AI automation projects. One of the key benefits of using claude-wt is its ability to streamline complex workflows. By managing multiple instances concurrently, users can significantly reduce the time spent on repetitive tasks and focus on more strategic activities. Although the exact time savings are currently unknown, the skill is built to facilitate faster execution of AI tasks, which can lead to improved project timelines and quicker iterations. This skill is particularly beneficial for developers, product managers, and AI practitioners who are looking to enhance their AI automation capabilities. It fits seamlessly into AI-first workflows, allowing teams to efficiently manage multiple projects without the overhead of manual coordination. For instance, a data engineering team could leverage claude-wt to run parallel data processing tasks, thereby accelerating data pipeline execution. With an intermediate implementation difficulty, users can expect to spend around 30 minutes setting up the claude-wt skill. While it does not require extensive prior knowledge, familiarity with Claude Code and basic automation principles will be advantageous. As organizations increasingly adopt AI automation, integrating tools like claude-wt into their workflows can provide a significant competitive edge, enabling teams to operate more effectively in a rapidly evolving digital landscape.
[{"step":"Install claude-wt via npm: `npm install -g claude-wt`","tip":"Ensure you have Node.js 16+ installed. Verify installation with `claude-wt --version`"},{"step":"Define your parallel tasks in a JSON config file or via command line. Example CLI: `claude-wt --instances 3 --command \"grep -r 'TODO' /codebase/\"`","tip":"Use `--timeout` flag to prevent hanging instances (e.g., `--timeout 300` for 5 minutes)"},{"step":"Execute the parallel commands and review aggregated output. For complex workflows, use the `--output-dir` flag to save individual instance logs.","tip":"For CPU-intensive tasks, match instance count to your CPU cores (e.g., `--instances 8` on an 8-core machine)"},{"step":"Process the results. Combine logs with `cat output/*.log > combined_results.txt` or use the JSON output for programmatic analysis.","tip":"Use `--cleanup` flag to automatically kill instances after completion if running in a shared environment"}]
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/jlowin/claude-wtCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Use claude-wt to manage parallel Claude Code instances for [TASK]. Execute [COMMAND_1], [COMMAND_2], and [COMMAND_3] simultaneously across [NUMBER] instances. Monitor progress and aggregate results into a single report. Example: 'Use claude-wt to run a parallel file search for "*.py" in /projects/[PROJECT_NAME] across 4 instances, then combine the output into a list of unique files.'
``` claude-wt --instances 4 --command "find /projects/ecommerce-backend -name '*.py' -type f" --output-format json Parallel Execution Summary: - Instance 1 (Port 4000): Completed in 8.2s (127 files found) - Instance 2 (Port 4001): Completed in 7.9s (134 files found) - Instance 3 (Port 4002): Completed in 8.5s (122 files found) - Instance 4 (Port 4003): Completed in 7.7s (141 files found) Aggregated Results: Total unique Python files: 524 Top directories by file count: /projects/ecommerce-backend/models (89 files) /projects/ecommerce-backend/services (76 files) /projects/ecommerce-backend/utils (62 files) Execution completed in 9.1s (parallel speedup: 3.4x vs sequential) ```
AI assistant built for thoughtful, nuanced conversation
AI sales agent for lead generation and follow-up
AI-assisted web application security testing
Get more done every day with Microsoft Teams – powered by AI
Agentic AI Workflow platform
Connected workspace for docs, wikis, and projects
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan