claude-cli-runner

Execute Claude Code CLI subprocesses with timeouts, budget guards, and stream-json logs.

1|1|Updated Jul 24, 2014
One-click install
npx skills add https://github.com/rmanzoku/dotfiles --skill claude-cli-runner-rmanzoku
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: claude-cli-runner
Source: https://github.com/rmanzoku/dotfiles/tree/main/skills/claude-cli-runner
Command: npx skills add https://github.com/rmanzoku/dotfiles --skill claude-cli-runner-rmanzoku

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires claude, and includes scripts (resource) components.

What problem does it solve?

This Skill simplifies the invocation of Claude Code CLI subprocesses, ensuring that long-running tasks are observable and manageable, with features like timeouts, budget guards, and artifact-based failure handling.

Core Features & Use Cases

  • Observable Execution: Offers detailed stream-json logs for every subprocess, making long-running tasks transparent.
  • Timeouts and Budgets: Ensures subprocesses do not run indefinitely, with configurable timeout limits and optional budget guards.
  • Artifact-based Failure Handling: Automatically manages failure scenarios by creating artifacts that provide insights into the task's outcome.
  • Use Case: When Codex needs to invoke a claude -p command for research, review, generation, or other CLI work, this Skill ensures that the process is not only executed but also observable and accountable.

Quick Start

To run a Claude Code CLI subprocess with stream-json logging and failure handling, use the following command:

run_claude_cli.py --prompt-file .context/<task>/prompt.md --output-dir .context/<task> --expected-artifact result.md

Frequently Asked Questions about claude-cli-runner

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run Claude Code CLI subprocesses with logging and timeout enforcement?

You can execute Claude Code CLI subprocesses with detailed stream-json logging, configurable timeouts, and budget guards using a Python 3 script that requires the 'claude' binary. This ensures long-running tasks remain observable and manageable.

What is the best way to handle failures when automating Claude Code CLI tasks?

The best way to handle Claude Code CLI failures is through artifact-based failure handling. The script automatically manages failure scenarios by creating artifacts that provide insights into the task's outcome.

How do I prevent a Claude Code CLI subprocess from running indefinitely?

To prevent a Claude Code CLI subprocess from running indefinitely, you can enforce configurable timeout limits and optional budget guards. This ensures processes are accountable and terminate within expected constraints.

Do I need Python to execute Claude Code CLI commands with budget guards and logging?

Yes, you need Python 3 and the 'claude' binary installed. The script uses Python to manage subprocess execution, enforce timeouts, guard budgets, and generate stream-json logs for observability.

How do I make long-running Claude Code CLI tasks observable?

You make long-running Claude Code CLI tasks observable by generating detailed stream-json logs for every subprocess. This logging mechanism makes task execution transparent and easy to monitor.