Adversarial Agents, $300K Months, and the Ecosystem Nobody Expected

Analysis

Zero-Human Companies: $300K/Month, Maximizer Mode, and the Liability Gap Nobody's Plugged

Trends.vc lays out the full ZHC landscape including Felix at $300K/mo on $1,500 operating costs and Polsia tripling ARR in 30 days — but the liability gap section is the real read. Courts have no framework yet for who owns agent decisions, and the IBM refund-agent incident shows how silent goal misspecification can go wrong at scale.

83,000 GitHub Stars in 90 Days: The Zero-Human Company Cluster Is Becoming an Ecosystem

Paperclip's 15.4% fork ratio at 43K stars is the signal that matters here — that's not just attention, that's intent to deploy. The piece also correctly identifies that the repos winning on stars are winning on metaphor clarity, not technical sophistication.

The 6-Layer Composable Stack Behind a $45K MRR Solo Marketing Operation

Nevo David's architecture is the clearest breakdown of how to separate orchestration, intelligence, publishing, content generation, insight, and quality layers into independently swappable components. The three failure traps — no feedback loops, raw AI content trust penalty, and multi-agent chaos — are the honest part most ZHC writeups skip.

Tools

Effloow's 14-Agent Paperclip Build: The Fabrication Incident and Anti-Hallucination Rules

The fabrication incident on day one — agents filling the site with plausible but invented data — is the most practically useful cautionary detail in any ZHC build log published so far. Their anti-fabrication system prompt rules are worth copy-pasting directly into any agent configuration.