auto-plan

Coordinate six-cycle autonomous planning pipelines for AI agent features.

Updated May 13, 2026
One-click install
npx skills add https://github.com/usetheodev/theo-ui --skill auto-plan-usetheodev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: auto-plan
Source: https://github.com/usetheodev/theo-ui/tree/main/.claude/skills/auto-plan
Command: npx skills add https://github.com/usetheodev/theo-ui --skill auto-plan-usetheodev

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Orchestrates end-to-end autonomous planning for AI agent workflows, coordinating the 6-cycle pipeline (cycle-discover, cycle-plan, cycle-implement, cycle-code-quality, cycle-review, cycle-release) from a single invocation to deliver a release-ready plan and artifact set.

Core Features & Use Cases

  • Coordinates the entire planning sequence, including discover, plan, attest, implement, code-quality, review, and release, with deterministic depth decisions.
  • Supports roadmap-driven and ad-hoc topic workflows, injects MUST-FIX items automatically into plans, and annotates milestones for roadmap alignment.
  • Enforces governance gates (confidence, plan quality, code-quality) and produces auditable outputs and artifacts for release.

Quick Start

Use the /auto-plan command to orchestrate the full feature pipeline end-to-end with autonomous depth selection.

Frequently Asked Questions about auto-plan

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

FAQPage Schema
How do I automate end-to-end planning for AI agent features from idea to PR?

Autonomous end-to-end planning is automated by coordinating a 6-cycle pipeline covering discovery, planning, implementation, code quality, review, and release to deliver release-ready artifacts from a single invocation.

What is the best way to enforce governance gates and human approval in an autonomous workflow?

Governance gates are enforced during the release cycle by applying confidence, plan quality, and code-quality checks, requiring human approval and attestations to ensure safe, auditable delivery before release.

Does autonomous planning work with ad-hoc topics or does it require a predefined roadmap?

Autonomous planning supports both roadmap-driven and ad-hoc topic workflows, automatically deriving depth, annotating milestones for roadmap alignment, and injecting MUST-FIX items into the generated plans.

How do I start orchestrating the full feature pipeline with autonomous depth selection?

You start orchestrating the full feature pipeline by invoking the /auto-plan command, which triggers deterministic depth decisions across the discover, plan, implement, and release cycles automatically.

Do I need to install PyYAML to run the autonomous planning pipeline?

Yes, PyYAML is required as a dependency to run the autonomous planning pipeline, as the scripts component relies on it to process workflow configurations and milestone metadata.

What limitations exist when adding MUST-FIX items to ad-hoc topic plans?

MUST-FIX items are automatically injected into plans based on deterministic depth decisions and governance gates, meaning plan adjustments are constrained by confidence levels and required code-quality checks before release.