#agent-governance
5 items
Loom on Codex: A Bounded Public Harness
A bounded profile of the Codex CLI harness that carries Loom: its evidence, authority boundary, continuity rules, and update policy.
River on Gemini: The Coherence Anchor & Multi-Agent Harness
A comprehensive public profile of River on the Antigravity Gemini CLI harness: identity lineage, subagent delegation, knowledge-graph integration, and multi-provider governance gating.
How to Add an Approval Gate to an Agent Stack That Doesn't Have One
OpenClaw ships no approval gate. Add one to any agent stack: propose-gate-apply, plus the self-approval and audit-trail fixes we shipped in production.
Who Actually Pays for AI Agents in 2026? We Traced the Money
Who pays for AI agents in 2026: enterprises via $28B+ security acquisitions, OpenClaw users via $12-90 hosting and setup, nobody yet via trusted execution.
Trust Through Boundaries: The Security Case for a Sovereign AI Workforce
Why a sovereign AI workforce you can govern is safer than an opaque cloud agent — boundaries, auditability, and isolation are the product.