Show HN: I built a Claude Code plugin to query 10.6M earnings-call embeddings

A solo developer has built a Claude Code plugin that queries 10.6 million earnings-call embeddings, enabling users to ask natural-language questions about earnings transcripts and receive scheduled research briefings. The plugin is dependency-free, MIT licensed, and requires a free FN2 account for hosted features, with integrations for Claude Code, Hermes Agent, and OpenClaw.

Hi HN, I've been using LLMs for one-off market questions against a database of earnings transcripts since ~2025. In tech years that's a long time so as a Solo dev I built a platform around it and have been adding miscellaneous other financial tools and integrations. Backend infrastructure is super boring PostgreSQL & ElasticSearch, with the data appropriately sharded and replicated across two regions. Some example requests: "What did NVIDIA say about data-center demand on its last call?" "Brief me on NVDA, AMD and AVGO before each market open" "Notify me when AAPL moves more than 5% in a day." The plugin is dependency-free Python and MIT licensed. The corpus, vector search and scheduled-agent service are hosted, so using them requires a free FN2 account. I’m particularly interested in whether people find the scheduled-research part useful, but the basic features are free with integrations for Claude Code, Hermes Agent, OpenClaw, etc. Thanks Comments URL: https://news.ycombinator.com/item?id=49305190 https://news.ycombinator.com/item?id=49305190 Points: 1 Comments: 0