rterakedis

rterakedis/leanwheel-skills

Plugin Claude Code

Leanwheel agile workflow skills for Claude Code.

rterakedis/leanwheel-skills

Plugin Claude Code

Leanwheel agile workflow skills: product discovery, PRD, epics, stories, UX, architecture, dev workflow, code review, retrospectives, and GitHub tracking.

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Author the App Store Connect inputs for an Apple app as lint-clean, committed artifacts under docs/store/ — framed screenshots at exact store sizes (ASSETS), listing copy per locale (METADATA), and a subscription-group / IAP spec reconciled against the project's .storekit and code (PRODUCTS). Use when the user says…

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Audit an iOS/iPadOS SwiftUI project's configuration, privacy declarations, and review-sensitive behaviors against App Store / TestFlight submission requirements. Produces a remediation story plus a non-code submission checklist. Use when the user says "app store preflight", "appstore preflight", "app store readiness"…

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Create or update the technical architecture document. Use when the user says "create architecture" or "update architecture".

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Validate that PRD, architecture, and epics are aligned and complete before starting implementation. Use when the user says "check readiness", "are we ready to build", or "validate planning docs".

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Review code changes adversarially. Use when the user says "review", "code review", or "review story {X}". Also embedded automatically at the end of /dev-story — see that skill for the inline variant.

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Manage mid-sprint changes and pull deferred items forward into the active workstream. Use when the user says "correct course", "change of plans", "we need to adjust", or "something came up".

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Create a comprehensive story file for a specific epic/story. Use when the user says "create story" or "create next story" or "create story {epic}-{story}".

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Composable operations on docs/project/decisions.md — the project-wide decision ledger. Called by ideate, spec, correct-course, and retrospective when a direction-setting decision is made, reversed, or parked. Directly invocable as /decision-log to view the current log and open questions.

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Composable deferred-item operations. Called by other skills when a finding is deferred. Directly invocable as /deferred to view the current log.

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Visually verify implemented UI against docs/ux/DESIGN.md tokens and EXPERIENCE.md states by building and observing the running app or site. Called by /dev-story for UI stories; directly invocable as /design-verify to check the current working tree.

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Implement a single stated goal in any folder — no project docs required. Grills for intent, writes a verifiable plan, executes, and verifies by running. Use when the user states a task/goal in a folder with no leanwheel docs/ tree, or says "single goal", "dev goal", "just do this here", or "one-off task".

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Implement a story from its story file. Use when the user says "dev story", "implement story", or "dev {story file path}".

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Reverse-engineer a brownfield project into docs/prd.md, docs/architecture.md, and CLAUDE.md. Use when the user says "discover project", "document this codebase", or is starting a new AI-assisted workflow on an existing project.

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Editorial review of a doc as writing — structure (cuts, reorganization), prose (clarity fixes), and an adversarial pass (gaps, ambiguities, contradictions). Use when the user says "doc review", "tighten this doc", "review the PRD as writing", or "editorial review".

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Composable documentation-maintenance operations — keep the human standup/ops/db guides and the canonical architecture doc current as code changes. Called by dev-story and quick-dev (OPERATIONAL), epic-flywheel and retrospective (PROMOTE), and code-review (DRIFT). Directly invocable as /docs-sync to catch the current…

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Generate automated API/E2E tests for features that already exist — brownfield code, features shipped before the eval net, or manual test-plan scenarios worth automating. Use when the user says "generate e2e tests", "backfill tests", "automate the test plan", or "qa tests for {feature}".

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Composable elicitation engine — pulls one decision at a time out of the user through sharp questioning and adversarial pressure. Called by ideate, spec, and correct-course; not normally invoked directly.

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Composable operations that keep docs/epics.md from growing without bound — collapse a closed epic's story detail to a summary row, and cut a shipped release into an archive file. Called by /retrospective. Directly invocable as /epic-archive.

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Drive an entire epic to completion semi-autonomously — runs the per-story flywheel (create → dev → review) for every story with granular commits, then a single epic-boundary verification + manual-test-planning pass. Use when the user says "epic flywheel", "run the whole epic", "flywheel the epic", or "run epic {N}…

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Break the PRD into epics and stories, write docs/epics.md, and create GitHub milestones. Use when the user says "create epics", "break down the PRD", or "plan stories".

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Composable eval operations — a persistent, stack-agnostic regression net plus optional LM-judge scoring. Called by create-story (BUILD), dev-story and code-review (RUN/SCORE). Directly invocable as /evals to run the current eval set. Works for Swift, web, and Python alike.

rterakedis/leanwheel-skills

Skill Claude CodeCodex

Adversarially pressure-test a formed idea via persona cross-examination until it Hardens, gets Killed, or becomes Clearer. Use when the user wants to stress-test, pressure-test, poke holes in, or "forge" an idea — invocable standalone or after writing a brief.