Weekly Winners

openTag is an open source AI coworker that lives in Slack and Lark. Tag it in a thread with a task, and a coding agent on your own computer does the work and replies there. It is built by first-tree.ai under the Apache 2.0 licence. There is no AI provider lock-in: bring a Claude or Codex subscription or any API key. Key Features Chat-first delegation: Tag openTag in any Slack or Lark thread and the task starts without leaving the conversation. Bring your own model: Run Claude Code, Codex, or Pi with your own subscription or API key, with no provider lock-in. Local execution: A small program on your Mac or Linux machine runs the agent, and openTag servers do not store your code by default. Context Tree memory: Team memory is markdown in a git repository inside your own GitHub organization, so it stays readable and versioned. 1000+ connectors: Agents reach GitHub, Notion, Linear, Jira, and Figma using your own credentials and minimal scope. Team mode: One agent serves a whole channel, and agents can hand work to each other and check progress. Who Should Use openTag Engineering teams that want issue triage, PR reviews, and status updates handled in team chat. Startups and solo maintainers who need extra hands without adding headcount. Privacy-conscious organizations that need agent memory to stay in their own GitHub organization. Frequently Asked Questions What is openTag? openTag is an open source AI coworker for Slack and Lark. You tag it in a thread, it runs a coding agent on your computer, and it replies with the result. Is openTag free? Yes. openTag is free to use, and you only pay for your own AI usage through your existing provider subscription or API key. Which apps does openTag support? It connects to Slack and Lark, and works with Claude Code, Codex, Pi, or any API key. Teams and Discord support is on the roadmap. Where does team memory live? By default openTag servers do not store your conversations, code, or memory. Memory lives in a git repository you host in your own GitHub organization.

LaunchGP is a launch platform for new products that runs every launch like a weekly race. Each Monday at 00:00 UTC, 32 products line up on the grid. Signed-in users vote once per car, and those votes set the running order on the live timing tower. Makers get attention, points and a dofollow backlink instead of another quiet directory listing. Key Features Weekly Grand Prix: a new grid of 32 products opens every Monday, so each launch gets one focused week in front of voters. Live timing tower: votes work as lap times and set the running order, with the top 10 scoring 25 points down to 1. Dofollow backlinks: every free car gets a dofollow link during its race week, and the top 3 keep theirs permanently. LaunchGP Endurance: every finisher joins an endless race where distance grows daily from real traffic, votes and shipping updates. Free and paid seats: 20 of the 32 seats are free with a badge on your site, while paid P1 to P6 seats run from $9 to $59. Who Should Use LaunchGP Indie makers and solo founders who want a launch with a real deadline, a public leaderboard and outside votes. SaaS and AI product teams hunting for early users, feedback and a permanent search link. Startup marketers who value dofollow backlinks and an ongoing race that keeps a product visible after launch week. Frequently Asked Questions What is LaunchGP? LaunchGP is a weekly launch race for new products. Products enter the grid, people vote, and season standings decide the podium. Is LaunchGP free? Yes. 20 of the 32 seats are free for makers who place the LaunchGP badge on their site, and each product gets one free Grand Prix per season. How do I get a backlink from LaunchGP? Every free car gets a dofollow link during race week. The top 3 of each Grand Prix keep a permanent link, and so do the top 10 of each Endurance class. What are the paid grid slots? Paid slots put you at the front of the grid for the week you pick, from $9 for P6 up to $59 for P1. Money buys visibility only, never votes or points.