run-plan

Discover pending work items and plan execution scope from state.yaml.

1|Updated Feb 13, 2026
One-click install
npx skills add https://github.com/phra/Antigravity-Discord-Bridge --skill run-plan-phra
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: run-plan
Source: https://github.com/phra/Antigravity-Discord-Bridge/tree/main/.specsmd/fire/agents/builder/skills/run-plan
Command: npx skills add https://github.com/phra/Antigravity-Discord-Bridge --skill run-plan-phra

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you plan the scope of your next execution run by intelligently discovering available work items and suggesting efficient groupings, ensuring you only execute what's necessary.

Core Features & Use Cases

  • Automated Discovery: Scans the file system and state.yaml to find all pending work items and intents.
  • Smart Grouping: Suggests optimal ways to group work items (single, batch, wide) based on dependencies, mode, and user history.
  • Dependency Management: Clearly defines that dependencies mean sequential execution within the same run, not separate runs.
  • User Choice: Always presents options and lets the user decide the execution scope.
  • Learning: Learns from user choices to improve future recommendations.
  • Use Case: Before starting a complex development task, you can use this Skill to see all the decomposed work items, understand their dependencies, and choose whether to run them one by one, grouped by type, or all together in a single, sequential run.

Quick Start

Use the run-plan skill to discover and plan the scope of pending work items.

Frequently Asked Questions about run-plan

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

FAQPage Schema
How do I plan the execution scope for pending work items?

To plan execution scope, you can scan the file system and state.yaml to discover pending work items, analyze their dependencies, and select an optimal single, batched, or wide grouping for your run.

What is the best way to group tasks with dependencies for a single run?

Grouping tasks with dependencies for a single run requires analyzing their relationships to ensure sequential execution within the same run, rather than triggering separate runs for each dependent item.

How does automated task discovery work for development planning?

Automated task discovery works by scanning your file system and state.yaml to identify all pending work items and intents, then presenting intelligent grouping options based on mode and user history.

Can I batch multiple work items together based on their dependencies?

Yes, you can batch multiple work items together by allowing the planner to analyze dependencies and suggest optimal batch groupings, while always retaining the final choice of execution scope to the user.

Does the execution planning tool learn my task grouping preferences over time?

Yes, the execution planning tool learns from your selected task groupings over time, using your historical choices to improve and refine future recommendations for pending work items.

What's the difference between single, batched, and wide execution modes for work items?

Single, batched, and wide execution modes differ by scale: single runs one item, batched groups related items by dependencies, and wide executes all pending work items together in a single sequential run.