Giving AI Agents a Soul: The Science Behind Persona ModelingWhen we started building Soul Spec, the thesis was simple: AI agents need identity files, not just system prompts. Give an agent a structured persona — personality, values, communication style — and it behaves more consistently, more safely, and more...Apr 17 · 3 min readclawsouls.hashnode.dev
Soul Spec v0.6: One Markdown File Is All You NeedWhen we released Soul Spec v0.3 two months ago, creating a persona required a soul.json with over ten mandatory fields, plus a SOUL.md, plus knowing the difference between specVersion and version. It worked, but we kept hearing the same thing: "I jus...Apr 13 · 3 min readclawsouls.hashnode.dev
Your AI Agent Needs an Approval System — Here Is How We Built OneAutonomous AI agents can now write code, deploy services, delete records, and send messages — all without a human touching a keyboard. That's the promise. It's also the risk. What happens when your agent decides to delete a database backup? Or push a...Apr 11 · 6 min readclawsouls.hashnode.dev
AI Doesn't Need a Bigger Engine. It Needs a Seatbelt.The 3/10 Problem Here's where AI adoption actually stands in most organizations: 3 out of 10 people use AI tools. The other 7 could, but don't. Not because the tools aren't impressive — they are. But because the answer to "what happens when it goes w...Apr 6 · 5 min readclawsouls.hashnode.dev
The Forest Has Parasites: Why AI Agent Security Needs Runtime Defense250 Documents. That's All It Takes. Last week, Anthropic published a joint study with the UK AI Safety Institute and the Alan Turing Institute that should make every AI developer uncomfortable: As few as 250 malicious documents can produce a backdoo...Apr 6 · 5 min readclawsouls.hashnode.dev
