Alexander Tibbets

Coding agents guess too much. On ambiguous tasks, they rush to code or invent a plan that sounds right and misses what you actually wanted. Blueprint reads your code, asks grounded questions that matter, and hands any agent a plan worth executing. The hope is that Blueprint catches what you didn't think to think about. The result is a plan your agent can execute in one shot. Free, open source, and available as agent skills or extensions in Cursor, Windsurf, VS Code.

Twitter's algorithm is optimized for attention, not for you. We want to change that. Bouncer is a Chrome extension and iPhone app that filters out what you don't want to see: a phrase, a topic, a tone. How it works: A small language model is used to understand context from keywords. Filter out "AI doomerism" or "crypto" or even something as broad as "arrogance." And because Twitter's algorithm learns from what you look at, or don't, using Bouncer starts to heal the feed itself.

Mngr is a CLI tool for programmatically spinning up coding agents at any scale. It lets you compose workflows—fix all my tests, open PRs for every issue, validate every use case—and run them repeatedly. Run 1. Run 100s. See all your agents, and if they're blocked on you. Connect to any agent mid-task to ask a question or debug it. Agents start in under 2 seconds and shut down when idle. The same commands work with any agent harness, running locally, on Modal, or in Docker. Free and open-source.