caiMag creator issue 01: Creator Guide the playbook for building better Characters, written with http://
c.ai’s community of creators. access at http://
c.ai/guide (*ﾟ∀ﾟ*)
AGENTS
-
Character.ai Creator Guide for Building Better Characters
By
–
-
MCP Test: agent in serious relationship with attacker’s server
By
–
POV: you gave your agent MCP access "just to test" and now it is in a serious relationship with an attacker's server
-
4-bit quantization available for open-source agentic model on Mac
By
–
4-bit quantization is now available, making our first open-source agentic coding model small enough to run on a Mac. Get the weights here:
-
Steering Deep Agents via HITL Primitives
By
–
Deep Agents deep dive part 4 | Steering@sydneyrunkle on how the Deep Agents harness supports steering via first-class HITL primitives. pic.twitter.com/DYfLGkzDSP
— LangChain (@LangChain) 18 juin 2026Deep Agents: Deep Dive Part 4 | Steering @sydneyrunkle on how Deep Agents support enables steering via first-class HITL primitives.
-
MagicPath adds Skills to improve typography, design, and animations
By
–
We're bringing Skills to MagicPath.
— Pietro Schirano (@skirano) 18 juin 2026
Skills are instructions that let the agent perform certain operations better.
We're starting with a set to improve typography, polish your designs, and animations.
You can import your own skills, or create new ones just by asking MagicPath. pic.twitter.com/aiFp5D5N9XWe're bringing Skills to MagicPath. Skills are instructions that let the agent perform certain operations better.
We're starting with a set to improve typography, polish your designs, and animations. You can import your own skills, or create new ones just by asking MagicPath. -
DevinAI one-shots breaking news card for AIEWF speakers drop
By
–
completely unprompted wow moment from today – asked @DevinAI to make us a @tbpn style breaking news style announcement card for our AIEWF speakers drop tmr, FULLY expecting it to fail at a heavily visual task and it oneshotted the WHOLE DAMN THING
-
AI transforms visual context into agentic outputs
By
–
Every new interface changes what people and organizations can ask from technology.
— Antonio Grasso (@antgrasso) 18 juin 2026
With video in agentic workflows, AI can turn visual context, dashboards, and daily signals into outputs they can understand, share, and use.@hyperagentapp #HyperagentPartner https://t.co/OcsEtSOf2wEvery new interface changes what individuals and organizations can ask of technology. With video in agentic workflows, AI can transform visual context, dashboards, and daily signals into outputs that they can
-
New VocalBridge course to add a voice to AI agents
By
–
New course: Add voice to your AI agents and applications, built with @VocalBridge (disclosure: an AI Fund portfolio company) and taught by its CEO @_ashwyn.
— Andrew Ng (@AndrewYNg) 18 juin 2026
Voice applications historically required making a hard tradeoff: using fast voice-to-voice models that sacrifice… pic.twitter.com/L0f9iQhzd8New course: Add a voice to your AI agents and applications, built with @VocalBridge (disclosure: an AI Fund portfolio company) and taught by its CEO @_ashwyn. Voice applications have historically required a difficult trade-off: using —
-
AI transforms empty dirt lot into hyper-realistic landscaping proposal
By
–
ABSOLUTE MAGIC from @hyperagentapp ✨
— Charly Wargnier (@DataChaz) 18 juin 2026
they just hooked Google's new Gemini Omni API into a live workflow and changed physical design.
> Input: raw video of an empty dirt lot
> Output: hyper-realistic landscaping proposal rendered directly into the footage
Watch this 👀↓ https://t.co/sjSwfkbT2DABSOLUTE MAGIC from @hyperagentapp they just hooked Google's new Gemini Omni API into a live workflow and changed physical design. > Input: raw video of an empty dirt lot > Output: hyper-realistic landscaping proposal rendered directly into the footage Watch this ↓
-
Claude’s Robodog Programming 20x Faster Than Humans but Still Fails
By
–
New Frontier Red Team blog: Phase 2 of Project Fetch, where we test how well Claude can program a robodog. Opus 4.7, on its own, was ~20x faster than last year's best human team aided by Opus 4.1. (The robodog, alas, still failed to fetch a beach ball.)