development-task-planner

Plan development tasks into ordered, low-risk, reviewable implementation increments.

2|Updated Apr 29, 2026
One-click install
npx skills add https://github.com/siarhei-belavus/agent-public --skill development-task-planner-siarhei-belavus
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: development-task-planner
Source: https://github.com/siarhei-belavus/agent-public/tree/main/skills/planning/development-task-planner
Command: npx skills add https://github.com/siarhei-belavus/agent-public --skill development-task-planner-siarhei-belavus

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Ambiguous engineering work often leads to risky, hard-to-review changes and plans that fall apart when a different person or agent implements them. This Skill turns vague development tasks into concrete, self-contained implementation plans with explicit scope, sequencing, validation, and amendment rules. ## Core Features & Use Cases - Structured Planning Workflow: Guides scope definition, change-surface analysis, success criteria, and low-risk increment slicing for features, refactors, migrations, and bug fixes. - Anti-Ambiguity Rules: Forces explicit decisions on naming, contracts, failure behavior, and ownership so a fresh implementer can execute without chat history. - Specialized Checklists: Loads reference guides for stateful systems, human maintainability, review alignment, and plan amendments when relevant. - Use Case: You need to migrate a background job queue to a new persistence layer. The Skill produces an ordered plan covering state ownership, retry and recovery behavior, validation steps, and explicit stop conditions if the codebase contradicts the plan. ## Quick Start Use the development-task-planner skill to turn this feature request into a step-by-step implementation plan with risks, validation, and amendment triggers.

Frequently Asked Questions about development-task-planner

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

FAQPage Schema
How do I break a large coding task into safe implementation steps?▼

Slice the work into increments that keep the system working after each step, isolating schema changes from behavior changes. Define observable success criteria and validation for each step before ordering them by the real dependency graph.

How to write an implementation plan another developer or AI agent can follow?▼

Make the plan self-contained: state exact target names for types, fields, and files, define failure behavior explicitly, and avoid vague wording like "consider" or "where possible". Include stop conditions and an amendment path requiring human approval when the plan is contradicted by the codebase.

What should a migration or refactor plan include?▼

Include a short objective, assumptions, ordered steps with clear dependencies, key risks, a validation strategy, and amendment triggers. For stateful systems, also plan for restart, retry, idempotency, cancellation, and status truthfulness.

When should implementation stop instead of improvising?▼

Stop when the codebase contradicts the planned change point, a required contract does not exist, or a step is unsafe in the real data model. Surface the blocker with options and pros/cons, then wait for a human decision rather than deviating silently.

When is a REVIEW_NOTE appropriate instead of a plan amendment?▼

Use a REVIEW_NOTE for non-blocking better alternatives, localized suboptimal choices, or future cleanup suggestions that do not change execution. Never use it for blockers, contradictions, or design changes that alter sequencing or scope.