implementation-plan-authoring

Creates persistent, resumable software implementation plans with checkpoints, invariants, and validation gates.

3|Updated Aug 8, 2026
One-click install
npx skills add https://github.com/Jose-Polanco-Oxte/Echos-Live-Music-Visualizer --skill implementation-plan-authoring-jose-polanco-oxte
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: implementation-plan-authoring
Source: https://github.com/Jose-Polanco-Oxte/Echos-Live-Music-Visualizer/tree/main/.agents/skills/implementation-plan-authoring
Command: npx skills add https://github.com/Jose-Polanco-Oxte/Echos-Live-Music-Visualizer --skill implementation-plan-authoring-jose-polanco-oxte

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Complex engineering work often loses critical context between sessions, forcing agents or developers to rediscover architectural decisions after interruptions. This Skill produces implementation plans that act as persistent execution contracts, preserving verified context, decisions, progress state, and validation criteria so another agent can resume work without reconstructing hidden reasoning. ## Core Features & Use Cases - Structured Plan Lifecycle: Manages plans through DRAFT, READY, IN_PROGRESS, BLOCKED, REPLANNING, and terminal states with a defined storage hierarchy under docs/public/plans and an INDEX.md catalog. - Execution Contract Definition: Captures requirements with traceability IDs, design decisions, invariants, change maps, dependency ordering, and explicit executor discretion boundaries. - Living Execution State: Maintains Current State Snapshots, checkpoint ledgers, compliance matrices, decision logs, deviation records, and handoff snapshots for safe resumption after context loss. - Use Case: When asked to plan a multi-milestone refactor such as an audio engine lifecycle migration, the Skill inspects repository evidence, records the baseline revision, and authors a READY plan with step-level validation gates that a separate execution agent can follow. ## Quick Start Ask the assistant to author a formal implementation plan for your feature or refactor before any code changes begin.

Frequently Asked Questions about implementation-plan-authoring

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

FAQPage Schema
How do I create an implementation plan that another agent can execute?

Request plan authoring explicitly and the Skill inspects repository evidence, records a baseline revision, and produces a structured plan with steps, invariants, validation gates, and a handoff snapshot. The plan is stored under docs/public/plans/active and indexed in INDEX.md.

When should I use plan authoring instead of coding directly?

Use it when the user explicitly asks for a plan, plan formalization, or replanning before implementation. The Skill is opt-in and does not activate merely because a task is complex or long-running.

Does the plan authoring skill also implement the code?

No. Authoring ends when the plan reaches READY status with an initial checkpoint. Implementation is delegated to a separate plan-conformance-execution skill after explicit execution authorization.

How are interrupted or long-running plans resumed?

Each plan maintains a Current State Snapshot, checkpoint ledger, and Handoff Snapshot recording the last safe checkpoint, next exact action, and pending validation. A resuming agent reads these sections first instead of reconstructing history.

What happens when a plan becomes outdated or invalid?

The plan transitions to REPLANNING, and if replaced, the old plan is marked SUPERSEDED with cross-references and moved to archive/superseded. Replan triggers such as invalid assumptions or API changes are defined explicitly in the plan.