execute-plan

Execute YAML-defined plans across sessions with phase-based state tracking.

1|Updated Apr 6, 2026
One-click install
npx skills add https://github.com/Honigbart/honeyflow --skill execute-plan-honigbart
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: execute-plan
Source: https://github.com/Honigbart/honeyflow/tree/main/execute-plan
Command: npx skills add https://github.com/Honigbart/honeyflow --skill execute-plan-honigbart

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Orchestrate long-running plan execution across sessions by advancing a finalized plan through its phases while preserving execution state and history.

Core Features & Use Cases

  • Phase-based execution: drives implementation from the active plan through each phase in order.
  • Stateful progress: records progress in .ai/plans/<slug>/execution_state.md and logs sessions in .ai/plans/<slug>/session_log.md.
  • Guardrails: enforces prerequisites, phase ordering, and review readiness before archiving.
  • Cross-plan coordination: supports multiple plans while preventing conflicts through structured state management.

Quick Start

Load the active plan's final_plan.md and begin executing the first phase, updating the execution_state.md as progress is made.

Frequently Asked Questions about execute-plan

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

FAQPage Schema
How do I execute a plan across multiple sessions while tracking progress?

Plan execution across multiple sessions is coordinated by advancing a finalized plan through ordered phases while recording progress in an execution_state.md file. This ensures stateful tracking and durable recovery for long-running implementation workflows.

What is phase-based plan execution and when do I need it?

Phase-based plan execution drives implementation through each phase in strict sequence, verifying prerequisites before advancing. You need this mechanism when orchestrating long-running, cross-session implementation tasks that require durable state preservation and recovery.

How do I start executing a finalized plan from a YAML definition?

To start executing a finalized plan, load the plan's final_plan.md file from the .ai/plans/<slug>/ directory and begin processing the first phase. The workflow automatically initializes the execution_state.md file to track ongoing versioned advancement.

Can I coordinate multiple plans simultaneously without causing conflicts?

Yes, cross-plan coordination supports multiple plans simultaneously by preventing conflicts through structured state management. Each plan maintains its own isolated execution_state.md and session logs within its respective .ai/plans/<slug>/ directory.

What happens if my plan execution is interrupted during a phase?

If plan execution is interrupted during a phase, the workflow preserves recovery by committing durable state changes to the execution_state.md file. This allows the next session to resume implementation exactly where the previous session left off.

Are there guardrails for phase ordering when advancing a plan?

Yes, the workflow enforces strict phase ordering, verifies prerequisites against active plan content, and checks review readiness before archiving. These guardrails prevent out-of-sequence execution and ensure all phase requirements are met.