auto-mode

Manage Claude Code execution permission levels across interactive sessions.

Updated Apr 28, 2026
One-click install
npx skills add https://github.com/laboratoriolegalcontable-png/stack-ia-creador --skill auto-mode-laboratoriolegalcontable-png
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: auto-mode
Source: https://github.com/laboratoriolegalcontable-png/stack-ia-creador/tree/main/.claude/skills/auto-mode
Command: npx skills add https://github.com/laboratoriolegalcontable-png/stack-ia-creador --skill auto-mode-laboratoriolegalcontable-png

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Auto Mode eliminates the constant interruptions caused by Claude Code repeatedly asking for permission on every step, so you can keep work moving without micromanagement.

Core Features & Use Cases

  • Three permission modes: Planeo (read-only), Automático (safe progress with confirmation where needed), and Omisión de permisos (run everything with minimal/no prompts in sandbox-like environments).
  • Multiple activation paths: Enable via app selector, cycle modes in terminal with Shift+Tab, or start with flags like --dangerously-skip-permissions in CI/terminal.
  • Safety-first workflow guidance: Recommends planning first in Planeo, then moving to Automático, and only using Omisión after repeated successful manual runs.

Quick Start

Tell Claude Code to enable Auto Mode by asking for “modo automático” or typing /auto-mode, then keep Planeo first and switch to Automático once the plan is approved.

Frequently Asked Questions about auto-mode

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

FAQPage Schema
How do I stop Claude Code from pausing for permission on every action in my terminal?

To stop Claude Code permission pauses, enable Auto Mode to manage execution levels across interactive sessions. It cycles through read-only planning, safe automatic execution, and full skip-permissions modes, reducing interruptions while maintaining control over your workflow.

Can I run Claude Code in CI without being prompted for manual approvals?

Yes, you can run Claude Code in CI without manual approvals by starting it with the --dangerously-skip-permissions flag. Auto Mode applies this Omisión de permisos level to execute everything without prompts, which is designed strictly for sandbox or worktree environments.

What are the different permission modes available for Claude Code automation?

The permission modes for Claude Code automation are Planeo for read-only planning, Automático for safe execution with confirmation where needed, and Omisión de permisos for running without prompts. You can cycle through these modes in the terminal using Shift+Tab.

How do I switch Claude Code workflow modes during an interactive session?

You switch Claude Code workflow modes during an interactive session by using the app selector or pressing Shift+Tab in the terminal. This cycles through Auto Mode's permission levels, allowing you to transition from planning to execution without leaving your session.

When should I use the dangerously skip permissions option in Claude Code?

You should use the dangerously skip permissions option only after repeated successful manual runs in sandbox-like environments or worktrees. Auto Mode recommends planning first in Planeo, moving to Automático for execution, and applying full permission skipping solely for isolated CI contexts.

Does Auto Mode work with developer productivity workflows involving code and documentation?

Yes, Auto Mode works with developer productivity workflows involving code, documentation, and CI contexts. It streamlines planning-to-execution workflows by managing permission levels, allowing you to progress through tasks with fewer terminal interruptions while retaining oversight.