08
Agent workflow: research-to-publish pipeline
A LangGraph workflow that finds trending topics, researches them, writes and illustrates a post, and publishes it to WordPress.
- Role
- Designed and built solo
- When
- 2026 · personal project, version 2 in progress
- Relevant for
- AI & AgentsSoftware
01 Finds trending topics worth writing about.
My part
I designed and built it on my own, from topic scoring to publishing.
Problem
Running a content site by hand takes hours per post: finding topics worth writing about, researching, writing, making images and publishing.
Approach
A LangGraph graph passes state between steps: topic discovery scored for demand, web research through Tavily, drafting with Azure OpenAI (GPT-5 mini/nano), a pgvector check so the site doesn't repeat itself, image generation with Gemini, then publishing through the WordPress API.
Outcome
Version 2 in progress.
What I did
A stateful graph
LangGraph passes state from one step to the next, from topic to published post.
LangGraphPython
Topics scored for demand
Trending topics are found and scored before anything is written.
Web research
Research on the chosen topic through Tavily.
Tavily
Drafting
The post is written with Azure OpenAI from the research.
Azure OpenAI (GPT-5 mini/nano)
No repeats
A pgvector check so the site doesn't repeat itself.
PostgreSQL + pgvector
Images and publishing
Images generated with Gemini, then the post goes out through the WordPress REST API.
GeminiWordPress REST API
Stack
- Python
- LangGraph
- Azure OpenAI (GPT-5 mini/nano)
- PostgreSQL + pgvector
- Tavily
- Gemini
- WordPress REST API