process-plan-executor

Executes written implementation plans in batches with review checkpoints between batches.

1|Updated Jun 23, 2026
One-click install
npx skills add https://github.com/bitranox/bitranox-skills --skill process-plan-executor-bitranox
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: process-plan-executor
Source: https://github.com/bitranox/bitranox-skills/tree/main/plugins/bitranox/skills/process-plan-executor
Command: npx skills add https://github.com/bitranox/bitranox-skills --skill process-plan-executor-bitranox

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Executing a written implementation plan in a fresh session often drifts: steps get skipped, verifications are forgotten, and nobody reviews the work until everything is done. This Skill enforces a disciplined loop of loading the plan, executing tasks in small batches, and pausing for architect review between batches. ## Core Features & Use Cases - Critical plan review before execution: Reads the plan, raises concerns with the human partner before writing any code, and tracks tasks via a todo list. - Batched execution with checkpoints: Implements the first three tasks by default, runs the plan's verifications, then reports results and waits for feedback before continuing. - Subagent model gate: Arms a hook during execution so any helper subagent dispatched mid-plan must pin its model explicitly, and disarms it on completion. - Use Case: You have a plan produced by a plan-writing skill and want it implemented in a separate session. Invoke this Skill, and it works through the plan batch by batch, stopping at blockers instead of guessing, and finishing with a green full test suite. ## Quick Start Use the process-plan-executor skill to implement the plan in docs/plans/feature-x.md, executing three tasks at a time and reporting back for review between batches.

Frequently Asked Questions about process-plan-executor

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

FAQPage Schema
How do I execute an implementation plan with Claude Code?▼

Load the plan file, review it critically and raise concerns first, then execute tasks in batches of three with a todo list. After each batch, show what was implemented and the verification output, then wait for feedback before continuing.

How many tasks should be executed before a review checkpoint?▼

The default batch is the first three tasks of the plan. Each task is marked in progress, its steps are followed exactly, verifications are run, and it is marked completed before reporting the batch for feedback.

What should I do when plan execution hits a blocker?▼

Stop executing immediately when you hit a missing dependency, failing test, unclear instruction, or repeated verification failure. Ask for clarification rather than guessing, and return to plan review if the fundamental approach needs rethinking.

What is the subagent model gate during plan execution?▼

It is a hook armed at execution start that denies any subagent dispatched during the plan unless it pins its model explicitly. The gate is disarmed when all batches are complete and the test suite is green.

When should I use subagent-driven development instead of plan execution?▼

Use subagent-driven development when you want the same plan implemented within the current session via implementer and reviewer subagents. Use plan execution when the work should happen in a separate session with human review checkpoints between batches.