Ring:Running Dev Cycle Frontend logo

Ring:Running Dev Cycle Frontend

Organization
LerianStudio
ring:running-dev-cycle-frontend

Running the frontend (React/Next.js/TS) dev cycle from a plan.md (ring:writing-plans format; legacy tasks.md only for in-flight cycles) or backend handoff: drives frontend agents through Gate 0 TDD plus accessibility/visual/E2E/perf checks, Gate 7 parallel review, and Gate 8 user validation, with rolling-wave phase boundaries. Use when starting or resuming a gated frontend dev cycle. Skip for backend (use ring:running-dev-cycle) or docs-only work.

Overview

PublisherLerianStudio
Repositoryring
Skill namering:running-dev-cycle-frontend
Stars
215
Forks
28
Bundled files
Instructions only
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by LerianStudio on GitHub. Read the source before you install it.

Installation

Install the Ring:Running Dev Cycle Frontend AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/LerianStudio/ring.git /tmp/ring
mkdir -p .claude/skills
cp -r /tmp/ring/dev-team/skills/running-dev-cycle-frontend .claude/skills/lerianstudio-ring-running-dev-cycle-frontend
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Ring:Running Dev Cycle Frontend in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Ring:Running Dev Cycle Frontend on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Ring:Running Dev Cycle Frontend is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Frontend Development Cycle Orchestrator

When to use

  • Starting a new frontend development cycle with a plan file (plan.md, canonical ring:writing-plans format; legacy tasks.md is accepted ONLY for cycles already in flight — current-cycle.json exists, init is not re-run)
  • Resuming an interrupted frontend development cycle (--resume flag)
  • After backend dev cycle completes (consuming handoff)

Skip when

  • No plan file exists
  • Task is documentation-only or planning-only
  • Backend project — use ring:running-dev-cycle instead

Sequence

Runs before: ring:writing-dev-reports

You orchestrate. Agents execute. NEVER read/write/edit source files (*.ts, *.tsx, *.jsx, *.css) directly. All code changes go through Task(subagent_type=...). Announce at start: "Using ring:running-dev-cycle-frontend with lean gate flow (Gate 0, 7, 8)."

Step 0: Pre-Execution Setup (MANDATORY)

1. Detect UI library: Read package.json
   <!-- Replace @your-org/design-system with your organization's design system package. -->
   - "@your-org/design-system" present → ui_library_mode = "design-system"
   - Otherwise → ui_library_mode = "fallback-only"
   Store in state.

2. Pre-cache standards (once):
   WebFetch → https://raw.githubusercontent.com/LerianStudio/ring/main/CLAUDE.md
   WebFetch → https://raw.githubusercontent.com/LerianStudio/ring/main/dev-team/docs/standards/frontend.md
   WebFetch → testing-accessibility.md, testing-visual.md, testing-e2e.md, testing-performance.md, devops.md, sre.md
   Store in state.cached_standards.

3. Load backend handoff if available: docs/ring:running-dev-cycle/handoff-frontend.json

4. Verify PROJECT_RULES.md exists → STOP if missing.

5. Ask execution mode: automatic | manual_per_epic | manual_per_task

Gate Map

GateCadenceSkillAgentPurpose
0taskring:implementing-tasksring:frontend / ring:ui-engineer / ring:bff-tsTDD, coverage, accessibility, visual/E2E/perf checks, local runtime
0.5task (conditional)ring:applying-composition-patternsring:frontendComposition refactoring when complexity signals detected
7epicring:reviewing-code9 defaults + triggered specialists via ring:reviewing-codeCode review
8taskring:validating-acceptance-criteriaUserAcceptance sign-off

All listed gates are MANDATORY. No exceptions.

Gate Agent Selection (Gate 0)

ConditionAgent
React/Next.js componentring:frontend
Design system UIring:ui-engineer
BFF / API aggregationring:bff-ts
Mixedfrontend first, then bff-ts

Pass ui_library_mode to every Gate 0 agent.

Frontend TDD Policy

Component LayerTDD Required?When
Custom hooksYES — RED→GREENGate 0
Form validationYES — RED→GREENGate 0
State managementYES — RED→GREENGate 0
Conditional renderingYES — RED→GREENGate 0
API integrationYES — RED→GREENGate 0
Layout / stylingNO — test-afterGate 0 visual checks
AnimationsNO — test-afterGate 0 visual checks
Static presentationalNO — test-afterGate 0 visual checks

Execution Order

yaml
for each epic:
  for each task:
    Gate 0
    [checkpoint if manual_per_task]

  # epic-level (after all tasks)
  Gate 7

  # task-level validation after review passes
  for each task:
    Gate 8
    Skill("ring:committing-changes")  # commit task work after Gate 8 user approval

  [checkpoint if manual_per_epic]

  # phase boundary — fires once, after the last epic of the current phase
  [if epic is last in its phase: Phase Cadence (see below)]

Phase Boundary (Rolling Wave)

Phases group epics and are elaborated one at a time. After the last epic of the current phase completes its Gate 0/7/8 flow, fire the phase boundary exactly once:

1. Close the finished phase in the plan: set its `## Phase Overview` Status cell →
   `Complete` (Edit on the plan file; skip silently if the table is absent —
   FALLBACK single-phase plan).
2. Checkpoint with the user: summarize the completed phase (epics done, review/validation
   outcomes) and confirm intent to continue.
2.5. Ask the user: "Open a PR for this phase?" → if yes: `Skill("ring:opening-pull-requests")` (optional, never automatic).
3. Elaborate the next phase's tasks inline under each epic as `#### Task N.M.T:`
   blocks, following the ring:writing-plans Task Format (`- [ ] Done` checkbox
   immediately under the heading, then Context, Implementation vision, Files,
   Verification, Done when). Detail exactly one phase ahead — never further.
4. Set the newly elaborated phase's Phase Overview Status cell → `Detailed`.
5. Set state.current_phase to the next phase and resume execution from its first epic.

Epic **Status:** lifecycle writes (same contract as ring:running-dev-cycle): the plan's epic **Status:** line is the write target throughout the epic loop — PendingDoing before the epic's first Gate 0, DoingDone after the epic passes Gate 7/8 and its checkpoint, DoingFailed on a hard block. Edit the plan file at each of these transitions, alongside the state write.

Do not elaborate more than one phase ahead — detail decays before execution reaches it.

Gate Execution Workflow (MANDATORY for every gate)

1. Skill("[sub-skill-name]")
2. Follow sub-skill dispatch rules
3. Task(subagent_type=...)
4. Validate output
5. Write state
6. Next gate

Sub-skill MUST be loaded before dispatching the agent.

Gate 0.5: Composition Complexity Scan (Conditional)

After Gate 0 passes, scan changed .tsx/.jsx files for composition complexity signals. If no signals are detected, skip this gate entirely (zero overhead).

Detection Heuristics

SignalThresholdDetection
Boolean prop count>=3 boolean props in Props type/interfaceGrep: prop?: boolean or prop: boolean
File size + hooks>200 lines AND >3 useState/useEffectLine count + hook grep
Conditional branches>=3 ternaries or && chains tied to boolean propsPattern: {isX && ...} or isX ? ... : ...

When Triggers Hit

  1. Skill("ring:applying-composition-patterns") — load the composition patterns skill
  2. Dispatch ring:frontend with flagged files + skill content as context
  3. Agent applies patterns in priority order:
    • 1.1 Eliminate boolean prop proliferation (CRITICAL)
    • 1.2 Extract compound components if applicable
    • 2.x Lift state only if warranted
    • 3-4 Apply only if natural fit
  4. Re-run tests — all MUST pass; coverage MUST NOT decrease vs Gate 0
  5. Commit via ring:committing-changes — expected message: refactor(component): apply composition patterns to <ComponentName>

Safety

  • If refactoring breaks tests irrecoverably → rollback to Gate 0 state, skip Gate 0.5, proceed to Gate 7 with note: "Composition refactoring attempted but reverted — review flagged files manually"
  • Coverage after refactoring MUST be >= Gate 0 coverage
  • Behavior MUST NOT change — refactor structure only (same props in = same render out)

Gate 7: Reviewers

Invoke Skill("ring:reviewing-code"). The ring:reviewing-code skill dispatches its 9 default reviewers plus triggered conditional specialists in parallel and applies its own pass/fail rules.

Gate Completion Criteria

GateRequired for COMPLETE
0TDD RED captured (behavioral) + GREEN passes; visual: implementation complete
0.5Conditional: no complexity signals OR refactored + tests pass + coverage >= Gate 0
7ring:reviewing-code PASS (all 9 defaults and triggered specialists)
8Explicit "APPROVED" from user

Former Gates 1-6 checks are owned by Gate 0 implementation and local verification.

State Management

State: docs/ring:running-dev-cycle-frontend/current-cycle.json (state schema v2.0.0).

Write after EVERY gate. If write fails → STOP.

json
{
  "schema_version": "2.0.0",
  "ui_library_mode": "",
  "tasks_file": "",
  "execution_mode": "",
  "current_gate": 0,
  "phases": [],
  "current_phase": "",
  "phase_checkpoint": {},
  "epics": [],
  "current_epic_index": 0,
  "current_task_index": 0,
  "gates_completed": {},
  "cached_standards": {}
}

Each entry in epics[] (Epic N.M) carries its own tasks[] array (Task N.M.T). Gate 0/8 run at task cadence over epics[current_epic_index].tasks[current_task_index]; Gate 7 runs at epic cadence over the union of that epic's tasks.

Blocker Handling

BlockerAction
Gate failureSTOP. Fix before proceeding.
Missing PROJECT_RULES.mdSTOP. Create using template.
Standards WebFetch failsSTOP. Report.
Architectural decision neededSTOP. Present options to user.

Frequently asked questions

What does the Ring:Running Dev Cycle Frontend AI skill do?

Running the frontend (React/Next.js/TS) dev cycle from a plan.md (ring:writing-plans format; legacy tasks.md only for in-flight cycles) or backend handoff: drives frontend agents through Gate 0 TDD plus accessibility/visual/E2E/perf checks, Gate 7 parallel review, and Gate 8 user validation, with rolling-wave phase boundaries. Use when starting or resuming a gated frontend dev cycle. Skip for backend (use ring:running-dev-cycle) or docs-only work.

Why use Ring:Running Dev Cycle Frontend on TypingMind?

Because you install it once and use it with any model. Ring:Running Dev Cycle Frontend is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Ring:Running Dev Cycle Frontend in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/LerianStudio/ring/tree/main/dev-team/skills/running-dev-cycle-frontend. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Ring:Running Dev Cycle Frontend?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Ring:Running Dev Cycle Frontend?

As many as you like. As long as a model supports skills, you can use Ring:Running Dev Cycle Frontend with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Ring:Running Dev Cycle Frontend AI skill free?

Yes. It is published on GitHub by LerianStudio under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇