Inspiration
The realization of how important yet overlooked local politics is compared to national politics.
What it does
CivicLens automatically discovers city agendas, extracts and summarizes them with AI, and delivers plain-language civic briefs to residents in a personal feed.
How I built it
I designed a looped agent pipeline: Discovery -> Extraction -> Summarization -> Storage, using ADK/A2A to pass structured outputs between agents and scale across many documents.
Challenges I ran into
Parsing inconsistent government PDFs and filtering only relevant links pushed me to refine extraction logic and fallback strategies for summarization.
Accomplishments that I'm proud of
Getting the summarization agent to output the correct format of info.
What we learned
Applying loop agent construction to a unique problem.
What's next for CivicLens
The sky's the limit!!
Log in or sign up for Devpost to join the conversation.