
Social listening, better compaction and new models
Ask AdaL what people are saying online, and keep more of a long session when it compacts. Plus Edge for browser use, GPT-6 Sol and Luna, and Jev nudging models that stop early.
Let's have a journey on agents and models.

Ask AdaL what people are saying online, and keep more of a long session when it compacts. Plus Edge for browser use, GPT-6 Sol and Luna, and Jev nudging models that stop early.
Curated reads from the SylphAI team on agents, product releases, and the community building around production AI.

AdaL scored 92% on BU100. The result is a systems score, not a model score: closed-loop control, CDP transport, liveness vs readiness, and the remaining 8%.

skills.sh, backed by Vercel, ships an install command and a leaderboard. @skills ships a reference and skips the install step entirely. Same protocol underneath — SKILL.md — two different bets on how an agent should reach a capability.

In 5 minutes, you'll understand Skills, how they differ from MCPs and AGENTS.md, and discover how Skills let your AI automatically reuse workflows instead of starting from scratch every chat.

Loop engineering is clearly where we're heading. The missing piece is making it simple enough and affordable enough that every engineer can actually use it.

Browser use, deep research, and video generation are no longer modes you switch into. They're capabilities AdaL loads on its own when your task calls for them.

AdaL scored 92% on BU100. The result is a systems score, not a model score: closed-loop control, CDP transport, liveness vs readiness, and the remaining 8%.

skills.sh, backed by Vercel, ships an install command and a leaderboard. @skills ships a reference and skips the install step entirely. Same protocol underneath — SKILL.md — two different bets on how an agent should reach a capability.

Today's coding agents are excellent tools for humans in the loop, but real autonomy may require a layer above the agent, not more capability inside it.

In 5 minutes, you'll understand Skills, how they differ from MCPs and AGENTS.md, and discover how Skills let your AI automatically reuse workflows instead of starting from scratch every chat.

Loop engineering is clearly where we're heading. The missing piece is making it simple enough and affordable enough that every engineer can actually use it.