News to PRD

8-year internet product manager, now indie developer. Xiaohongshu: Handsome AI Evolution

News to PRD is an automated product ideation engine based on multi-agent collaboration. The system monitors tech news sources like Hacker News in real-time, automatically completing the full chain from 'one news item' to 'deployable MVP PRD' through a pipeline of 6 specialized AI Agents—News Hunter, Research Expert, Analyst, Product Insight Expert, PRD Expert. Core idea: Decompose PM workflow into orchestrable Agent collaboration network, achieving end-to-end automation of 'trend capture→market research→pain point mining→product definition'. Each news item creates independent project context, ensuring isolation and traceability during parallel processing. This isn't a simple AI writing tool, but a 24/7 product ideation factory, enabling indie developers and startup teams to continuously obtain high-quality product inspiration at minimal cost.

Overall Architecture: Event-driven multi-agent pipeline architecture, 6 specialized Agents orchestrated through Router, forming News Hunter → Web Searcher → Analyst → Product Insight → PRD Expert chain. Core Tech Stack: Agent Framework: OpenAgents SDK, supporting declarative YAML config and Python extensions. Communication Protocol: gRPC for efficient inter-Agent communication, HTTP/MCP for Studio visualization monitoring. Search Engine: Brave Search API primary, DuckDuckGo backup, ensuring research data reliability. LLM Backend: Pluggable design supporting OpenAI/Claude/local models. Key Technical Highlights: 1. Event-driven orchestration through project.notification.started → task.delegate → task.complete event chains for precise pipeline control, loose coupling between Agents. 2. Context isolation: Each news item creates independent Project with run_id tracking throughout, supporting parallel processing without interference.

Implemented Features (MVP Complete): News Scraping: Automatically fetches latest tech news from Hacker News, supports deduplication and rate control. Smart Research: Web Searcher automatically searches related info, returns structured evidence with source links. Market Analysis: Generates research reports with market trends, competitor analysis, user discussions, key insights. Product Insights: Mines target users, pain points, web solution elements. PRD Generation: Outputs complete MVP PRD with product name, requirements doc, SEO keywords (5-10), domain recommendations (3-5). Concurrency Control: Drop-on-Overload strategy, gracefully skips with notification when system is busy. Visual Monitoring: Studio UI for real-time pipeline status and output viewing. Planned Features: Multi-source access (Product Hunt, Reddit, Twitter/X), Smart summarization, PRD export to Notion/Markdown/PDF, Topic scoring, User subscriptions.