frontend-architecture-planner
Skill7 starsChanged 21 days agoApache-2.0View on GitHub
Plan architecture and ownership for standard or deep React/Next.js work across routing, state, data, styling, forms, builds, migrations, and implementation handoff. Skip micro-fixes and unrelated stacks.
In this repository23 other files it gives its agents.
- AGENTS.mdAGENTS.md
- agent-rules-skill-authorSkill
- agent-self-improvementSkill
- design-screenshot-specSkill
- execution-plan-managerSkill
- frontend-bugfix-debuggerSkill
- frontend-design-directorSkill
- frontend-design-intelligenceSkill
- frontend-layout-implementerSkill
- frontend-linter-managerSkill
- frontend-prototype-explorerSkill
- frontend-quality-reviewerSkill
- frontend-refactor-surgeonSkill
- frontend-visual-qaSkill
- goal-plannerSkill
- greenfield-project-builderSkill
- loop-workflow-plannerSkill
- mcp-toolchain-managerSkill
- pattern-library-managerSkill
- project-context-adapterSkill
- project-onboarding-adapterSkill
- webdev-feedbackSkill
- webdev-kit-updaterSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
19 rules. About 0 are common, 14 are distinctive, and 5 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈14
Distinctive · estimate
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 46
Use this skill after prompt intent routing when the task is Standard Workflow or Deep Workflow, the project is within the target stack, and …
- Distinctive: line 48
Use this skill when the user asks to:
- Distinctive: line 63
Do not use this skill for:
- Very specific: line 79
Read common/prompt-intent-routing-rules.md when workflow level is unclear.
- Distinctive: line 80
Read the compact goal contract or execution plan when present.
- Distinctive: line 81
Read only relevant boundary rules for the task.
- Distinctive: line 82
Read relevant project overlays when present: stack, architecture, styling, verification, docs, build, workspace, state, and data-fetching …
- Distinctive: line 83
Read affected source files only when architectural boundaries cannot be planned from overlays.
- Very specific: line 84
Do not read generated dist/ during normal runtime.
- Very specific: line 86
When React/Next work changes loading, imports, subscriptions or expensive rendering, read the relevant section of …
- Distinctive: line 95
Must not install packages, UI libraries, state libraries, data libraries, form libraries, or MCP servers.
- Distinctive: line 96
Must not change project source files, configs, package manager files, or build tooling.
- Distinctive: line 97
Must not interact with production systems, secrets, or production data.
- Distinctive: line 102
If the prompt is lightweight and no architecture boundary is affected, stop and route to the direct skill.
- Distinctive: line 108
Create the architecture handoff without editing source files.
- Very specific: line 146
If a material UI or state choice remains unresolved after targeted source and documentation inspection, hand off one bounded question to …
- Distinctive: line 156
Every proposed new boundary has a reason.
- Distinctive: line 157
No package, UI library, or MCP install was proposed as default.
- Distinctive: line 159
No source files, configs, or package files were changed by this skill.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Frontend Architecture Planner
Purpose
Plan frontend architecture decisions for React and Next.js projects when a task affects structure, ownership, routing, React state, Redux state, TanStack remote state, Axios data access, CSS Modules styling, form behavior, build/workspace shape, or long-lived conventions.
This skill creates an architecture decision and implementation handoff. It does not implement code by default.
When To Use
Use this skill after prompt intent routing when the task is Standard Workflow or Deep Workflow, the project is within the target stack, and architecture decisions matter.
Use this skill when the user asks to:
- plan a new React or Next.js feature that touches multiple modules;
- design a route, screen group, or application section;
- define component boundaries before implementation;
- decide where React, Redux, or TanStack state should live;
- decide how Axios or TanStack data fetching should fit the existing project;
- plan routing changes;
- plan form ownership and validation boundaries;
- plan a frontend refactor before editing code;
- assess architecture risk before implementation;
- define a greenfield React or Next.js architecture before scaffolding.
When Not To Use
Do not use this skill for:
- one small bug;
- one obvious type error;
- one small styling adjustment;
- one direct file-scope edit;
- unrelated frontend frameworks or app stacks;
- purely visual direction without architecture impact;
- rendered visual QA;
- UI component library selection;
- backend, database, ORM, migrations, or infrastructure work unless the user explicitly changes the scope.
Required Context
- Read
AGENTS.md. - Read
common/target-stack-policy.md. - Read
common/prompt-intent-routing-rules.mdwhen workflow level is unclear. - Read the compact goal contract or execution plan when present.
- Read only relevant boundary rules for the task.
- Read relevant project overlays when present: stack, architecture, styling, verification, docs, build, workspace, state, and data-fetching profiles.
- Read affected source files only when architectural boundaries cannot be planned from overlays.
- Do not read generated
dist/**during normal runtime.
When React/Next work changes loading, imports, subscriptions or expensive
rendering, read the relevant section of common/react-performance-rules.md.
Tool Contract
- May use filesystem access to inspect relevant project overlays and affected source files.
- May use
context7for React, Next.js, Redux, TanStack, Axios, TypeScript, or build-tool documentation when architecture depends on current behavior. - Activate
openai_platform_docsonly when current OpenAI API or ChatGPT Apps SDK behavior affects the architecture. - May use MDN for platform constraints that affect architecture, such as forms, browser APIs, accessibility, and CSS behavior.
- Must not install packages, UI libraries, state libraries, data libraries, form libraries, or MCP servers.
- Must not change project source files, configs, package manager files, or build tooling.
- Must not interact with production systems, secrets, or production data.
Workflow
- Confirm target stack scope. If the project is outside React, Next.js, CSS Modules, Redux, TanStack, or Axios scope, report unsupported scope.
- If the prompt is lightweight and no architecture boundary is affected, stop and route to the direct skill.
- State the architectural question.
- Inspect current conventions: React/Next.js router, folder/module structure, component ownership, CSS Modules, Redux ownership, TanStack ownership, Axios API layer, form conventions, build/workspace shape, and verification commands.
- Define only relevant boundaries: route/page, feature/module, shared vs local component, React local state, Redux object state, TanStack remote state, Axios adapter, form state, CSS Modules styling, accessibility, build/workspace, and verification.
- Choose the smallest architecture that satisfies the goal.
- Require explicit approval for package installation, framework migration, package manager changes, new state/data/form/styling libraries, global architecture layers, build tooling changes, large refactors, destructive operations, or production access.
- Create the architecture handoff without editing source files.
- Define verification strategy using existing project checks and rendered verification where relevant.
- Hand off to
execution-plan-manageror the relevant implementation skill.
Output Contract
Final response: return only facts that affect the user's understanding, confidence, or next action. Omit empty fields and workflow narration.
Return or write a Frontend Architecture Plan with:
Goal
Workflow Level
Target Stack Scope
Architecture Question
Current Project Conventions
Proposed Boundaries
Files Or Surfaces Likely To Change
React State Ownership
Redux Boundary
TanStack Boundary
Axios Boundary
Routing Boundary
Form Boundary
CSS Modules Boundary
Build Or Workspace Boundary
Approval Gates
Implementation Slices
Verification Strategy
Risks
Rejected Alternatives
Next Skill Or Next StepOmit irrelevant sections rather than filling them with generic text.
Unresolved Experimental Questions
If a material UI or state choice remains unresolved after targeted source and
documentation inspection, hand off one bounded question to
frontend-prototype-explorer. Keep this planning skill read-only. Return the
observed decision and its limitations to the existing plan or decision log.
Validation Gates
- The task actually needed architecture planning.
- The plan is inside the supported target stack or clearly reports unsupported scope.
- The plan follows current project conventions where known.
- Every proposed new boundary has a reason.
- No package, UI library, or MCP install was proposed as default.
- Approval gates are explicit.
- No source files, configs, or package files were changed by this skill.
Trigger Evals
Should trigger:
- "Plan the frontend architecture for this React analytics section."
- "Where should this feature live in the current Next.js project?"
- "Design the route and Redux/TanStack boundaries before implementation."
- "Plan a safe frontend refactor for this module."
- "Create the architecture plan for a greenfield Next.js app, but do not scaffold yet."
Should not trigger:
- "Fix this TypeScript error."
- "Change this button color."
- "Make this page look less generic."
- "Run visual QA."
- "Install shadcn."
Reference Map
common/react-performance-rules.mdAGENTS.md.common/target-stack-policy.md.common/prompt-intent-routing-rules.md.common/framework-adaptation-policy.md.common/state-ownership-rules.md.common/routing-boundary-rules.md.common/data-fetching-boundary-rules.md.common/form-boundary-rules.md.common/build-tool-boundary-rules.md.common/frontend-integration-boundaries.md.skills/goal-planner/SKILL.md.skills/execution-plan-manager/SKILL.md.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

