5/ If you like the direction lclq is going, please star the repo and share your feedback.

Release v0.2.0: github.com/erans/lclq/r...

1/ lclq v0.2.0 is out! Push Subscriptions are now supported. You can receive Pub/Sub messages directly to your HTTP endpoints with no polling.

Release notes: github.com/erans/lclq/r...

1/ 🚀 pgsqlite v0.0.19 is live! More PostgreSQL catalog support on top of SQLite, making PG clients behave even more naturally. Lightweight PG compatibility FTW.

github.com/erans/pgsqli...

1/ 🚀 New LunaRoute feature drop!

Introducing limits-alternative - a smart routing strategy that keeps your AI agents running even when you hit rate or subscription limits.

👉 Seamless failover across models, providers, or accounts.

github.com/erans/lunaro...

1/ One great thing about good AI agents is they can finally build the tools I never had time for. This week, one of them tackled something I’ve postponed for years: a fast, local alternative to Google’s Pub/Sub emulator → github.com/erans/lclq 🧵

1/ 🚀 Meet LunaRoute - observability for AI assistants.

🔗 github.com/erans/lunaro...

If you use OpenAI, Anthropic, or Claude Code… you’re flying blind.

You don’t really know what your assistant is doing, how many tokens it burns, or what tools it’s calling. 🧵

5/ Maybe RSS never really died. It just evolved.
We might be entering a new era where Markdown becomes the feed format for AIs.

eran.sandler.co.il/post/2025-10...

1/ Remember RSS, Atom, and FeedBurner? They promised a better web - one feed reader for everything you cared about. Google Reader ruled that world until social networks and mobile apps took over, and the open feed era quietly faded.

1/ Attended a hackathon at
@cursor.com.web.brid.gy
's offices last week.

Big thanks to the team for hosting 🙌

I built AutoAgent Action, a GitHub Action that runs AI-powered checks on your PRs using Cursor CLI or other AI CLIs. 🚀

Blog post: eran.sandler.co.il/post/2025-09...

5/ If you are working with an awkward API or one that is hard to wrap with an MCP server, consider creating a CLI around it. Expose clean commands, JSON output and scoped auth. Your API can become instantly human and LLM friendly. Read more on the blog:
eran.sandler.co.il/2025/09/02/w...

1/ Command line interfaces used to be for terminal experts and power users. Today something new is happening: large language models are becoming surprisingly good at wielding CLIs. Want to know how and why it matters? Read more: eran.sandler.co.il/2025/09/02/w... 🧵