write-plan

Convert design docs and findings into executable PLAN.md files with verifiable tasks.

3|Updated Oct 28, 2020
One-click install
npx skills add https://github.com/k0d3x8its/dotfiles --skill write-plan-k0d3x8its
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: write-plan
Source: https://github.com/k0d3x8its/dotfiles/tree/main/claude/.claude/skills/write-plan
Command: npx skills add https://github.com/k0d3x8its/dotfiles --skill write-plan-k0d3x8its

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Turning a resolved design document into an actionable, trackable implementation plan is error-prone: tasks get lost, verification steps are skipped, and downstream tools like Trello sync fail on malformed structure. This Skill converts a grilled design doc plus .work/FINDINGS.md into a properly structured .work/PLAN.md that /sync-trello and /dev-brief consume unchanged. ## Core Features & Use Cases - Structured plan generation: Emits Goals, Micro-Goals, and Tasks where every Task carries a verify sub-bullet — the command that proves the task is done. - Dual format support: Detects flat vs. index+detail plan layouts (.work/plan/ directory) and appends new Goals without clobbering existing ones or their Trello card IDs. - Conditional security gate: Offers a /threat-model design review only when the plan exposes an HTTP surface or handles sensitive data, then recommends /sync-trello at the end. - Use Case: After /grill-me resolves the open questions in a design doc for a new authenticated API, run /write-plan to produce a PLAN.md with FR-NN-cited tasks, get prompted for a threat-model review, and sync the result to your Trello board. ## Quick Start Run /write-plan with the path to your design doc to generate an executable .work/PLAN.md with verifiable tasks.

Frequently Asked Questions about write-plan

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

FAQPage Schema
How do I turn a design doc into an executable implementation plan?▼

Run /write-plan with the design doc path after resolving open questions with /grill-me. It reads the doc plus .work/FINDINGS.md and emits .work/PLAN.md with Goals, Micro-Goals, and Tasks, each carrying a verify command.

What plan format does /sync-trello parse for task tracking?▼

It parses a hierarchy of Goal headings, Micro-Goal subheadings, and checkbox task lines. Tasks must sit under a Micro-Goal, and verify sub-bullets are indented without checkboxes so they are never sent to Trello.

Can I add new goals to an existing PLAN.md without losing Trello card IDs?▼

Yes, the Skill appends new Goals only and never clobbers existing Goals or their [trello:ID] tags. This append-only rule applies to both flat PLAN.md files and the index+detail .work/plan/ layout.

When does write-plan recommend a threat model review?▼

It offers /threat-model only when the plan implies an HTTP or network-reachable surface, or handling of user, sensitive, or regulated data. Local-only CLIs, docs changes, and pure-logic libraries skip the offer silently.

What happens if my design doc still has open questions?▼

The Skill stops and recommends running /grill-me first. Every unanswered open question would otherwise become a wrong guess baked into the plan, so planning is blocked until decisions are resolved.