skills.rest
amoutiers
amoutiersCOMMUNITY
Explore all skills by this author →
GitHub

x-develop

Delegate plan execution to fresh subagents with structured review loops.

1|Updated May 28, 2026
Search Tags:#code-review#software-engineering#plan-execution#task-delegation#workflow-management#subagent-driven-development
One-click install
npx skills add https://github.com/amoutiers/maxi-superpowers --skill x-develop
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: x-develop
Source: https://github.com/amoutiers/maxi-superpowers/tree/main/skills/x-develop
Command: npx skills add https://github.com/amoutiers/maxi-superpowers --skill x-develop

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The skill provides a disciplined method to delegate plan execution to ephemeral implementer subagents, ensuring clean handoffs, fresh context, and auditable progress throughout a maxi plan.

Core Features & Use Cases

  • Fresh subagent per dispatch: always deploy a new implementer subagent for each task iteration to avoid context leakage.
  • Structured review loop: allow up to three fix-review cycles per task before escalation, preserving alignment with the spec.
  • Final whole-branch review: run a comprehensive cross-task review using the standard code-review template to surface regressions and interface mismatches.
  • Durable progress ledger: automatically record task states and outcomes to guarantee accountability.

Quick Start

Ask maxi to deploy a fresh implementer subagent and begin the specified task using the subagent-driven workflow.

Frequently Asked Questions about x-develop

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

FAQPage Schema
How do I delegate plan execution to subagents without context leakage?▼

Subagent-driven development delegates plan execution to fresh, transient implementer subagents for each task dispatch. This workflow ensures clean handoffs, fresh context, and auditable progress throughout the maxi plan without context leakage.

How does a structured review loop work for task delegation?▼

A structured review loop allows up to three fix-review cycles per task before escalation, preserving alignment with the spec. It automatically records task states and outcomes to a durable progress ledger for guaranteed accountability.

What is the best way to maintain cross-task consistency in software engineering workflows?▼

The best way to maintain cross-task consistency is to run a final whole-branch review using the standard code-review template. This comprehensive review surfaces regressions and interface mismatches across all dispatched tasks.

How do I track task outcomes and states during subagent-driven development?▼

You track task outcomes during subagent-driven development using a durable progress ledger. This ledger automatically records task states and native DONE or NEEDS_CONTEXT statuses to guarantee accountability throughout the workflow.

When should I escalate a task in a subagent-driven workflow?▼

You should escalate a task in a subagent-driven workflow after completing three fix-review cycles. The structured review loop caps iterations at three to preserve spec alignment before escalating the task for further context.

Related Skills

Explore More Software Engineering Skills→
earchibaldearchibald
70

subagent-driven-development

Dispatch fresh subagents per task with two-stage spec and code quality reviews.

Community
Advanced
amoutiersamoutiers
1

subagent-driven-development

Distribute implementation tasks across isolated subagents with per-task reviews.

Community
Advanced
shmily1012shmily1012
1

subagent-driven-development

Dispatch a fresh subagent per task with a two-stage review workflow.

Community
Advanced
islee23520islee23520
5

ulw-loop

Orchestrate durable work into evidence-backed execution loops with checkpointed progress.

Community
Advanced