Case Study: A Zero-Cost Release Notes Bot on a Free Model and a Free Server

Description and use

An automated bot generates weekly release notes using a free LLM and server, costing zero.

Problem

Release notes are often skipped, leading to changelog drift and confusion about what changed in a release.

Outcome

The bot runs weekly, consumes <10M tokens per run, and produces a draft changelog with accurate grouping, all on free infrastructure.

Workflow

• Fetch merged PRs from GitHub API for the last 7 days
• Build a concise prompt using only PR titles
• Send prompt to MonkeyCode free model endpoint
• Receive grouped release notes
• Write output to CHANGELOG.md

Evidence

Build a tight prompt using only PR titles

Source

The total infrastructure cost is $0

Source