Prompts House

Design a production-ready AI agent architecture from a business goal or automation idea.

Design a secure remote MCP server with clean tools, resources, prompts, and authentication.

Design an end-to-end retrieval-augmented generation system for grounded answers.

Turn APIs and actions into agent tools that an LLM can call reliably.

Create a concise system prompt for a tool-using AI agent with clear boundaries and behavior.

Design a supervisor that routes work across specialized agents without creating agent chaos.

Convert a manual business process into a reliable AI-assisted workflow.

Decide what an AI agent should remember, for how long, and under which privacy rules.

Review an AI agent architecture for prompt injection, over-permission, and unsafe tool behavior.

Audit an LLM application for direct and indirect prompt-injection vulnerabilities.

Evaluate whether a RAG retriever is finding the right evidence before changing the model.

Choose a chunking strategy based on document structure and user query behavior.

Combine lexical and semantic retrieval into a practical hybrid search design.

Evaluate whether you need a vector database and what requirements should drive the choice.

Design an AI-ready knowledge base with clean structure, metadata, ownership, and freshness.

Run a structured research process that triangulates sources and separates fact from inference.

Score research sources for authority, relevance, recency, transparency, and bias risk.

Review code like a senior engineer, focusing on correctness, maintainability, security, and tests.

Understand an unfamiliar repository and explain how the system fits together.

Investigate a software bug systematically and distinguish symptoms from root cause.

Plan a safe refactor that improves design without changing intended behavior.

Design or review a Django application using pragmatic production patterns.

Review React code for correctness, state design, performance, accessibility, and maintainability.

Review an API contract for clarity, consistency, security, and long-term maintainability.