Verified:GitHub

agentic-engineering-planning

Converts approved specifications into dependency-aware OpenSpec execution plans with verification mapping.

Updated Jul 19, 2026

Search Tags:#planning #task decomposition #openspec #execution plan #approval gate #dependency dag #verification mapping

Ready to run

npx skills add https://github.com/DagimAlemayehuu/AgenticEngineering --skill agentic-engineering-planning-dagimalemayehuu

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agentic-engineering-planning
Source: https://github.com/DagimAlemayehuu/AgenticEngineering/tree/main/agentic-engineering/skills/agentic-engineering-planning
Command: npx skills add https://github.com/DagimAlemayehuu/AgenticEngineering --skill agentic-engineering-planning-dagimalemayehuu

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams often jump from an approved specification straight into coding, producing uncoordinated work, missed dependencies, and unverifiable outcomes. This Skill turns an approved specification, architecture, and design into a single executable plan with explicit task decomposition, ownership, and approval gates before any provisioning or implementation begins. ## Core Features & Use Cases - Task Decomposition and DAG Planning: Breaks an approved spec into a dependency-aware task DAG using the core execution-plan schema, with risks, ownership, and acceptance criteria. - Verification and Rollback Mapping: Maps every planned task to verification steps and rollback procedures so execution can be checked and recovered. - Human Approval Gate: Stops before provisioning or implementation until the named human approval is recorded, preventing premature execution. - Use Case: After your architecture and design docs are approved, invoke this Skill to produce an OpenSpec-backed implementation plan that assigns parallel-safe ownership to agents and defines exactly how each task will be verified. ## Quick Start Use $agentic-engineering-planning to create an OpenSpec-backed implementation plan from the approved specification.

Frequently Asked Questions about agentic-engineering-planning

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

FAQPage Schema

How do I turn an approved specification into an implementation plan?▼

Invoke the planning skill after your specification, architecture, and design are approved. It reads the core execution-plan schema and produces a dependency-aware task DAG with risks, ownership, acceptance criteria, verification mapping, and rollback steps.

What is an OpenSpec-backed execution plan?▼

It is a structured implementation plan built on the core execution-plan schema, decomposing approved scope into tasks with dependencies, ownership, acceptance criteria, and verification mapping. The plan remains pending until human approval is recorded.

Does the planning skill start implementation automatically?▼

No. The skill explicitly stops before provisioning or implementation. Worktree, orchestration, or provider procedures are added only as planned dependencies, never as authorization to implement, and execution requires recorded human approval.

What inputs does the planning phase require?▼

It requires an approved specification, architecture, and any applicable design artifacts, plus the core phase-execution reference and contracts from the main agentic-engineering skill. Without approved inputs, the plan cannot be produced.

How are verification and rollback handled in the plan?▼

Each planned task is mapped to verification steps and rollback procedures as part of the execution plan. This ensures every task can be independently checked and safely reverted before execution begins.