Briefly: A Local-First, Zero-Data-Egress Filing Assistant for Legal Practice
This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend What I Built My wife is a practicing attorney managing an active case load. Like many legal practitioners working against tight court filing deadlines, her day-to-day workflow involves downloading court orders, disclosure bundles, witness statements, client emails, and vendor invoices directly into her local…
This weekend challenge submission focuses on a local-first filing assistant called Briefly, designed to aid legal practitioners with their document management workflow. Practicing attorneys often struggle with the time-consuming task of downloading case-related documents and then searching through numerous unstructured files in their Downloads folder to locate specific pleadings or affidavits before important hearings.
To address this issue, the Briefly assistant monitors a designated inbox directory on the user's machine, extracts text from supported document formats (PDF, DOCX, TXT, and Markdown), and utilizes an open-weight model to identify the corresponding legal matter and document category. Files that receive high-confidence matches are automatically filed into a standardized matter directory structure, while ambiguous, scanned, or low-confidence files remain untouched in the inbox for manual review.
The developer, AnthonyASBaptiste, refined the tool based on real-world feedback from a lawyer using the assistant. The initial confusion around interpreting the cooldown timer in seconds versus minutes led to the implementation of standardized minute presets (1 min, 5 min, 10 min, 15 min, 30 min, 1 hr) in the interface. The lawyer also expressed interest in searching for keywords across all documents within a matter, confirming that the matter hierarchy organization solved the foundational problem, even though cross-document indexing was out of scope for the current MVP.
Briefly is built using standard library dependencies only (Python 3.10+) and a local Ollama instance, which allows for open-weight model inference. The tool employs Google's open-weight Gemma 4 (gemma4:e2b-it-qat) for document classification and filing. Once the assistant completes the first-pass discovery and organizes the documents into clean matter folders, the lawyer can easily navigate the directory tree and search for specific keywords using the built-in search functionality.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
