maiconfz--plan-refiner--automatic-plan-refiner

Refines an existing draft plan against repository context with labeled assumptions in one pass.

3|1|Updated May 2, 2026
One-click install
npx skills add https://github.com/agents-repo/registry --skill maiconfz-plan-refiner-automatic-plan-refiner-agents-repo
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: maiconfz--plan-refiner--automatic-plan-refiner
Source: https://github.com/agents-repo/registry/tree/main/.cursor/skills/maiconfz/plan-refiner/maiconfz--plan-refiner--automatic-plan-refiner
Command: npx skills add https://github.com/agents-repo/registry --skill maiconfz-plan-refiner-automatic-plan-refiner-agents-repo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Draft plans often contain gaps, unverified file references, and hidden assumptions that surface only during implementation. This Skill performs an assumption-first quality pass on an existing plan, checking it against the actual repository so the plan is complete and feasible before any code is written. ## Core Features & Use Cases - Bounded repository rediscovery: Reads README, specs, rules, and agent instructions, verifies files the plan cites, and searches related modules and tests the planner missed. - Critique-then-revise workflow: Applies a structured review rubric covering goal alignment, completeness, consistency, feasibility, sequencing, test coverage, and failure modes, then produces a complete refined plan in one pass. - Labeled assumption logging: Fills gaps with conservative, documented assumptions recorded both in the refined plan and in a separate assumption log. - Use Case: You have a draft migration plan written by another tool. Hand it to this Skill with your goal statement, and receive a revised plan that cites real repository files, flags risks, and lists every assumption made. ## Quick Start Refine my draft plan in docs/plan.md against the goal of adding user authentication, and list any assumptions you make.

Frequently Asked Questions about maiconfz--plan-refiner--automatic-plan-refiner

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

FAQPage Schema
How do I refine an existing implementation plan with AI?▼

Provide the draft plan as markdown text or a workspace file path, optionally with a goal statement and prior clarifications. The Skill critiques the draft against the repository, fills gaps with labeled assumptions, and returns a complete refined plan in one pass.

What does the plan refinement review check?▼

The review rubric covers goal alignment, completeness, internal consistency between steps, files, and tests, feasibility against the current repository tree, scope boundaries, sequencing, test coverage, hidden assumptions, over-engineering, and failure modes for migrations or breaking changes.

Can it create a plan from scratch if no draft exists?▼

No. If the draft plan is empty or unusable, it leaves the refined plan empty and requests a usable draft through open questions. It is designed to revise existing plans, not to author a first plan or intake GitHub issues.

Does the plan refiner modify repository code?▼

No. It never implements product code, commits, pushes, or opens pull requests. The only file it may update is the source plan file itself when a known path exists and write-back applies.

What happens to assumptions made during plan refinement?▼

Every gap is filled with a conservative default and explicitly labeled. Assumptions appear both in an Assumptions section inside the refined plan markdown and in a separate assumption log output, so a copied plan stays self-contained.