
Build interactive AI agents in Elixir with OTP supervision, middleware composition, human-in-the-loop approvals, sub-agent delegation, and real-time Phoenix LiveView integration. Built on LangChain...
Wanna find garbage tests in your #ElixirLang codebase?
I'm working on a Credo check that will flag any test that fails to exercise production code. Stuff like:
test "to_string works" do
assert to_string(:user) == "user"
end
(That's a real example I found in our ~20k test codebase!)
🧵
While debugging my new coding agents orchestrating tool called Camelot today I developed more features for https://lamabot.social than during three previous months.
If you're using the #ElixirLang NodeJS package, be advised there's an information disclosure security vulnerability (CVE-2026-33872).
It'd be quite difficult for an attacker to intentionally exploit, but fairly easy to accidentally trigger yourself. Update to 3.1.4 ASAP.
https://github.com/revelrylabs/elixir-nodejs/security/advisories/GHSA-rwcr-rpcc-3g9m

