Guides Claude when building, debugging, or extending features in the AutoBidder Next.js 15 + FastAPI codebase. Use this skill whenever the user mentions adding a feature, fixing a bug, writing an API endpoint, creating a React component, working on the frontend or backend code, asking about project conventions, setting up Docker, managing environment variables, or doing anything related to AutoBidder's tech stack (Next.js, FastAPI, TanStack Query, shadcn/ui, TailwindCSS, PostgreSQL, JWT auth, Resend, Vercel, Railway). Also trigger when the user shares a file path, pastes code from the project, or asks "how does X work in this project". Always consult this skill before writing any AutoBidder code — it encodes the project's conventions and will prevent pattern drift.
Guides Claude through AutoBidder's end-to-end AI proposal generation workflow. Use this skill whenever the user wants to generate, draft, review, or refine a freelance proposal — including when they paste a job description, say "write a bid", "generate a proposal", "draft a proposal for this job", "help me respond to this posting", "what should I write for this job", or share a job URL. Also trigger when the user asks about proposal strategy, how to customize a template, or whether a job is a good fit. Even if the request seems simple ("quick proposal for this"), use this skill — it encodes AutoBidder's full quality workflow and should always be consulted when proposals are involved.
Guides Claude when managing, debugging, or improving AutoBidder's RAG knowledge base pipeline. Use this skill whenever the user mentions: "knowledge base", "RAG", "ChromaDB", "embeddings", "document upload", "retrieval", "the AI isn't using the right context", "portfolio docs", "LangChain", "chunking", "vector store", "bad proposals", or "why did it generate that". Also trigger when the user wants to upload new portfolio documents, inspect what's in the knowledge base, tune retrieval quality, debug why a proposal got irrelevant context, or rebuild/reset the vector store. Always consult this skill before touching anything in backend/app/services/knowledge_service.py or the ChromaDB collection — it encodes all ingestion, retrieval, and debugging patterns for this project.
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
Generate a custom checklist for the current feature based on user requirements.
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
Create or update the project constitution from interactive or provided principle inputs, ensuring all dependent templates stay in sync.
Execute the implementation plan by processing and executing all tasks defined in tasks.md