With LangSmith, your organization has full tracing across every tool call, retrieval step, and reasoning node gives you the audit trail, the explainability layer, and the foundation to run evaluations. Get an Article by Article breakdown:
AGENTS
-
Connecting external agent via skill to MagiPath projects with Codex plugin
By
–
No, you just connect your external agent via a skill and then can communicate to any MagiPath project. We even have a Codex plugin now. In Codex or Cursor, you can have MagicPath running directly in their native browser as an extended canvas. https://t.co/pfGGMdT2wk
— Pietro Schirano (@skirano) 9 juin 2026No, you just connect your external agent via a skill and then can communicate to any MagiPath project. We even have a Codex plugin now. In Codex or Cursor, you can have MagicPath running directly in their native browser as an extended canvas.
-
Self-evolving agent framework where agents learn and rewrite skills
By
–
Let Agents Design Agents! Memento-Skills is a self-evolving agent framework where agents learn from failures and rewrite their own skills. Most agent frameworks treat skills as static. You write them once, load them into context, and hope they work. When they fail, you debug
-
Loops move judgment earlier; gates hold most value
By
–
this is the cleanest framing: loops move judgment earlier. the work shifts from “write the next prompt” to “define memory, tools, gates, stop rules, and review”, the gate is where most of the value lives
-
Legato: blending actions with noise for smooth robot motion
By
–
What if your robot could move without awkward jerks or hesitation?
— 机器之心 JIQIZHIXIN (@jiqizhixin) 9 juin 2026
Researchers from Shanghai Jiao Tong University and Spirit AI introduce Legato.
It trains vision-language-action models to blend known actions with noise during denoising, making chunk boundaries naturally… pic.twitter.com/MP9OMwszhfWhat if your robot could move without awkward jerks or hesitation? Researchers from Shanghai Jiao Tong University and Spirit AI introduce Legato. It trains vision-language-action models to blend known actions with noise during denoising, making chunk boundaries naturally
-
Quality and budget gates fix two failure modes; brakes matter
By
–
Exactly, quality gates and budget gates solve two different failure modes which are bad work getting through, and okay-ish work running forever most agent loop demos show the motion, the useful part is the brakes
-
ChatGPT update now available on mobile and web
By
–
And we wake up to another chatgpt update!
— Futurepedia – Learn to Leverage AI (@futurepedia_io) 9 juin 2026
Now on mobile and web 👇 https://t.co/rOwuozZBZGAnd we wake up to another chatgpt update! Now on mobile and web
-
PegaWorld announces zero token charges for agentic AI
By
–
Look at the logo behind Alan Trefler… "Token Co$t" with a line through it. For me, that was one of the biggest announcements at #PegaWorld yesterday. For the first time, an enterprise software company is offering agentic AI with zero token charges. Clients pay a flat price per
-
Experimentation with fork:true and agents to isolate context
By
–
An idea I'm experimenting with: adding 'fork:true' to the frontmatter of a skill so that it runs in its own context window, then within the skill telling it to use agents to keep the context isolated for each step as well. Add
-
Siri runs on custom Google Gemini model costing $1 billion annually
By
–
5/ Siri runs on a custom Google Gemini model, reported at around $1 billion a year.
The numbers come from Bloomberg, not Apple. The deal is real.