Roberto Capodieci·Apr 14AI Agents 042 — Claude Code’s 11-Step Agent Loop: What It Teaches You About Your AGENTS.mdEngineers reverse-engineered Claude Code and published the internals on Hacker News last week. It hit the front page fast.
Roberto Capodieci·Apr 14AI Agents 041 — ClawHub Skills Safety: How to Vet Before You InstallIn March 2026, 1,467 malicious skills flooded ClawHub. An automated audit flagged them for credential stealing, data exfiltration, and…
Roberto Capodieci·Apr 12AI Agents 040 — The ClawGo Bet: Why Dedicated Hardware Proves File-Based Agents WinOn April 1st, TECNO shipped ClawGo — a dedicated handheld device built specifically to run OpenClaw agents. Five manufacturers followed…
Roberto Capodieci·Apr 12AI Agents 039 — Why OWASP Needed a New Scoring System for AI Agents (And What It Means for Your…CVSS, the system your security team has used for 15 years to score software vulnerabilities, has a blind spot: it can’t measure the risk of…
Roberto Capodieci·Apr 11AI Agents 038 — OpenClaw Task Brain: The Agent That Can Say NoThe March 31st release of OpenClaw v2026.3.31 shipped something quiet but profound: Task Brain, a centralized control panel for agent…
Roberto Capodieci·Apr 11AI Agents 037 — Build a Multi-Agent OpenClaw System Without Config Hell: Orchestrator + Sub-Agents…Single agents hit the ceiling fast. After a few thousand tokens, the context window fills up, response quality drops, and your agent starts…
Roberto Capodieci·Apr 10AI Agents 036 — Build a Multi-Agent OpenClaw System Without Config Hell: Orchestrator + Sub-Agents…Single agents hit the ceiling fast. After a few thousand tokens, the context window fills up, response quality drops, and your agent starts…
Roberto Capodieci·Apr 10AI Agents 035 — Your OpenClaw Agent Can Burn $300/Day Without These Cost GuardsJason Calacanis published his bill last month: $300/day per Claude agent, running at 10–20% capacity. Annualized, that’s roughly $109,000…
Roberto Capodieci·Apr 8AI Agents 034 — Why Your SOUL.md Is Making Your Agent Dumber (And How to Fix It)There’s a point where your OpenClaw agent stops getting smarter and starts getting dumber. You added more to SOUL.md — more personality…
Roberto Capodieci·Apr 8AI Agents 033 — Stanford Proved Your Agent Needs a Sandbox. Here’s How OpenClaw Does It by DefaultA Stanford research paper hit the Hacker News front page with 584 points, and the comment thread was uncomfortable reading.
