From solopreneur
Reviews technical plans against platform-specific best practices before implementation. Detects stacks like Swift/iOS, Kotlin/Android, React/Next.js, Python/FastAPI, LangGraph; queries docs via context7 and dispatches subagents.
npx claudepluginhub hanamizuki/solopreneur --plugin neo4j-devThis skill uses the workspace's default tool permissions.
Pre-implementation technical plan review. Verify whether the approach follows best
Reviews implementation Plan files in parallel using Codex, Gemini, and Claude to analyze validity, gaps, risks, and improvements. Invoke via /plan-review after plan creation.
Reviews execution plans for architecture patterns, tech debt implications, suboptimal technology choices, and scalability using cto-advisor skill.
Reviews implementation plans for parallelization potential, TDD adherence, type and API verification, library choices, and security before execution.
Share bugs, ideas, or general feedback.
Pre-implementation technical plan review. Verify whether the approach follows best practices and surface potential issues before any code is written.
Determine where the plan comes from:
todos/backlog/xxx.md, docs/spec/xxx.md) → read itIdentify keywords in the plan and map to subagents and context7 query targets:
| Keywords | Platform | Subagent | context7 Query Targets |
|---|---|---|---|
| Swift, SwiftUI, @Observable, SwiftData, iOS | iOS | ios-dev | SwiftUI, Swift concurrency, relevant Apple frameworks |
| Kotlin, Compose, Room, ViewModel, Android | Android | android-dev | Jetpack Compose, Kotlin Coroutines, relevant Jetpack libraries |
| React, Next.js, TypeScript, TSX | Web | nextjs-dev / web-dev | React, Next.js |
| FastAPI, Python, Pydantic, CRUD, REST API | Backend (general) | python-dev | FastAPI, Pydantic |
| LangGraph, prompt, agent, tool calling, streaming, RAG, embedding, chain, node, state graph, LLM | Backend (LLM) | llm-dev | LangGraph, LangChain |
If multiple platforms are involved → run Step 3 + Step 4 for each.
Based on specific APIs or frameworks mentioned in the plan, query official documentation.
Query flow:
mcp__context7__resolve-library-id to find library IDs (e.g., "SwiftUI", "Jetpack Compose")mcp__context7__query-docs to look up specific API usage from the plan (e.g., "SwiftData relationship cascade delete", "Compose LazyColumn performance")Skip context7 when:
Based on detected platforms, dispatch corresponding subagents in parallel.
Subagent selection with fallback:
Use the platform-specific subagent type from the table above (e.g., ios-dev, python-dev).
If the preferred subagent type is not available in the current environment, fall back to
general-purpose with the same prompt — the review will be less specialized but still useful.
Subagent prompt template:
You are an expert reviewer, not an implementer. Do not write code or modify any files.
Review the following technical plan and answer three questions:
1. Does the approach follow best practices for this platform? If not, point out
what's wrong and suggest alternatives.
2. Are there existing codebase patterns to reference? (Search relevant files to confirm)
3. Are there potential pitfalls or common mistakes during implementation?
Plan content:
[paste full plan]
Official docs summary (context7 results):
[paste Step 3 results, or omit if nothing was queried]
Check the skill index for relevant best practices first, then provide your analysis.
Return analysis only — no code.
Combine subagent results and output in the following format:
[List detected platforms and corresponding subagents]
[One section per platform, listing: what's good ✅, what's problematic ⚠️, suggested adjustments 🔧]
[List implementation details that are easy to miss. Omit this section if there are none.]
[If adjustments are needed, give a specific direction for each, described in one sentence]
[One paragraph: is the approach viable overall? Does it need changes before implementation?]