https://capabilityhostprotocol.com/blog On governing and proving what AI agents and systems do. https://capabilityhostprotocol.com/blog/the-agentic-web-has-no-evidence-layer https://capabilityhostprotocol.com/blog/the-agentic-web-has-no-evidence-layer Wed, 24 Jun 2026 00:00:00 GMT We're building careful ways for agents to discover tools, call them, and prove who they are. We've built almost nothing to prove what they actually did. That missing layer is where trust between agents will either hold or break. https://capabilityhostprotocol.com/blog/introducing-capabilities-txt https://capabilityhostprotocol.com/blog/introducing-capabilities-txt Mon, 22 Jun 2026 00:00:00 GMT Agents can find what to read (llms.txt) and respect what to crawl (robots.txt). There's no standard way for a host to advertise what it can do. capabilities.txt is that missing layer. https://capabilityhostprotocol.com/blog/logs-arent-evidence https://capabilityhostprotocol.com/blog/logs-arent-evidence Thu, 18 Jun 2026 00:00:00 GMT When an AI agent does something consequential and someone asks what happened, scattered application logs aren't an answer. Here's the difference between a log and evidence — and why it's a protocol problem. https://capabilityhostprotocol.com/blog/why-a-protocol-not-a-feature https://capabilityhostprotocol.com/blog/why-a-protocol-not-a-feature Tue, 16 Jun 2026 00:00:00 GMT Every agent framework will add its own audit log. So why does governing AI actions need an open protocol? Because evidence you can trust has to outlive the system that produced it. https://capabilityhostprotocol.com/blog/proving-why-a-claim-was-denied https://capabilityhostprotocol.com/blog/proving-why-a-claim-was-denied Thu, 11 Jun 2026 00:00:00 GMT Insurers are automating approve/deny decisions faster than they can defend them. Denial is a first-class outcome in CHP — which makes 'show me why' a recorded fact, not a reconstruction. https://capabilityhostprotocol.com/blog/the-capability-boundary https://capabilityhostprotocol.com/blog/the-capability-boundary Sat, 06 Jun 2026 00:00:00 GMT You can't govern an AI agent by watching the model. Governance happens at the capability boundary — the moment an action crosses from intent into effect. Here's why that line is the right place to stand. https://capabilityhostprotocol.com/blog/chain-of-custody-for-ai-review https://capabilityhostprotocol.com/blog/chain-of-custody-for-ai-review Mon, 01 Jun 2026 00:00:00 GMT When AI reads, summarizes, and flags documents in legal review, the work product is only as defensible as its provenance. CHP's evidence is hash-chained — which is, almost literally, chain of custody. https://capabilityhostprotocol.com/blog/the-security-review-that-stalls-your-agent https://capabilityhostprotocol.com/blog/the-security-review-that-stalls-your-agent Tue, 26 May 2026 00:00:00 GMT You built the agent. It works. Then a security review asks what it did and whether it was allowed to — and the rollout stops. That gap is where CHP starts, and it's real today. https://capabilityhostprotocol.com/blog/controls-in-the-contract https://capabilityhostprotocol.com/blog/controls-in-the-contract Thu, 21 May 2026 00:00:00 GMT When AI sits inside trading, credit, and payments decisions, 'demonstrate the controls' is the question that arrives from model risk and regulators. CHP puts the controls in the capability contract — enforced before invocation, not asserted after. https://capabilityhostprotocol.com/blog/who-authorized-the-ai-step https://capabilityhostprotocol.com/blog/who-authorized-the-ai-step Sat, 16 May 2026 00:00:00 GMT AI scribes and agents draft notes, summarize charts, and prepare orders — but a clinician stays in command. CHP records the sign-off as a governed approval in the same trace, capturing who-did-what without storing the PHI body. https://capabilityhostprotocol.com/blog/who-commanded-the-machine https://capabilityhostprotocol.com/blog/who-commanded-the-machine Tue, 12 May 2026 00:00:00 GMT Work orders, dispatches, and machine commands are increasingly issued by software and agents, not just people. When something goes wrong, the question is who commanded it, whether it was approved, and whether a safety condition was checked. https://capabilityhostprotocol.com/blog/chp-and-mcp https://capabilityhostprotocol.com/blog/chp-and-mcp Wed, 06 May 2026 00:00:00 GMT The Model Context Protocol answers 'what can the model call.' CHP answers 'what actually happened, and can I prove it.' They're different layers of the same stack — here's how they fit. https://capabilityhostprotocol.com/blog/evidence-is-not-telemetry https://capabilityhostprotocol.com/blog/evidence-is-not-telemetry Thu, 30 Apr 2026 00:00:00 GMT You already run OpenTelemetry. So why would agent actions need a separate evidence layer? Because telemetry is built to help you understand a system — and evidence is built to be defended. Different jobs, and they compose.