Creates structured implementation plans with YAML frontmatter and Playwright verification steps.
npx skills add https://github.com/hpark0011/mirror --skill create-plan-hpark0011
Please help me install this Agent Skill. Skill: create-plan Source: https://github.com/hpark0011/mirror/tree/main/.agents/skills/create-plan Command: npx skills add https://github.com/hpark0011/mirror --skill create-plan-hpark0011
workspace/plans/ with date-and-slug filenames and a strict YAML frontmatter contract (id, slug, status, type, branch, worktree, verification tier). - Hard verification built in: Every plan must include a Playwright CLI test path with assertions, plus ordered implementation steps and explicit constraints and non-goals. - Traceability: Sequential PLAN_NNN ids, predecessor/successor links, and ticket references keep plans connected to the rest of the workflow. - Use Case: You receive a requirement to add a new feature to the Mirror app. Invoke this Skill to research the current state, allocate the next plan id, and produce a complete plan file with E2E verification steps, then open it in Greyboard for review. ## Quick Start Ask the assistant to plan this requirement and write an implementation plan for it before any code is written.High-intent search queries and answers about installing and using this skill.
Invoke the create-plan skill with your requirement. It researches the current state, allocates the next sequential plan id, and writes a plan file to workspace/plans/ with frontmatter, implementation steps, and a Playwright verification section.
Required fields are id, slug, title, date, type, status, branch, worktree, scope, apps, and verification_tier. Optional fields like predecessor, successor, and tickets are added only when those relationships exist.
Ids use the PLAN_NNN format with three-digit zero-padding, allocated sequentially. The skill scans existing plan contents with grep for the highest id in use, since filenames are date and slug based rather than id based.
Yes. Every plan must include a hard verification step: a Playwright CLI test path with assertions. Chrome MCP is allowed only for visual confirmation, not for test assertions.
Status starts as draft, flips to active when the branch work begins, becomes completed when merged, superseded when a later plan replaces it, and blocked only when an external constraint stops progress.
smartfrogAsk 1-5 essential clarifying questions before starting ambiguous work.
DNepovimExecutes written implementation plans task-by-task with verification checkpoints and review gates.
commercetoolsCreate Jira epics and child tickets from markdown or OpenSpec plans.
shermanhumanGenerate step-by-step implementation plans with files and verification commands.