Heso Self Serve - gate every agent action, prove it to anyone

Building agents is getting easier. Trusting them is not. Most systems can tell you what an agent *claims* it did. Heso gives you cryptographic proof of what actually happened. Every agent action can be gated by policy, require human approval when needed, and generate an Ed25519-signed receipt that anyone can verify independently. No dashboards to trust, no vendor lock-in, no "just trust our logs." We built Heso because AI agents are starting to move money, access systems, and make real decisions. When something goes wrong, editable logs aren't evidence. Happy to answer questions, hear feedback, and discuss where agent infrastructure is headed.