Defining the boundaries of autonomy.
Horonom builds infrastructure products for autonomous software systems — assembled, governed, and auditable. Software is becoming autonomous, but autonomy without boundaries becomes drift. We build the systems that make autonomy explicit, governable, and safe to scale.
Our namesake does not create the sky. It observes it, understands it, records it. When a new system is born, it links new stars — until the whole Intelligence Universe is charted. Every product we ship is a step in that charting: boundary-aware by design, composable rather than monolithic, and auditable from the start.
The current public product portfolio — see the Product Atlas
for the full system map and how these relate. Source of truth:
horonomy/official-website's
Product Registry
(HORO-282). This section is machine-generated from a pinned snapshot of that
Registry — see scripts/generate_profile_products.py (HORO-599). Do not
hand-edit between the markers; re-pin and regenerate instead.
Agent runtime & governance — Gives AI agents a runtime with permissions, approval checkpoints, and an audit trail instead of unrestricted tool access.
🌐 agent-assembly.com · 📚 docs · 💻 github.com/ai-agent-assembly
Change safety — Examines supported OpenAPI contract changes and proposes repairs for human review; an experimental static-analysis prototype.
🌐 octans.horonom.com · 📚 docs
Provenance & authority — Establishes who or what is authorized to take a sensitive action, and proves it after the fact.
🌐 circinus.horonom.com · 📚 docs
Context continuity — Carries context across machine, tool, and user boundaries so it is not re-derived or lost at each hop.
🌐 ophiuchus.horonom.com · 📚 docs
Agent integrity — Verifies real evidence for what an AI coding agent claims it did.
🌐 fornax.horonom.com · 📚 docs · 💻 github.com/horonomy/fornax-core
Product truth & integrity — Keeps what a product claims and what a product actually does from silently drifting apart.
🌐 horologium.horonom.com · 📚 docs
More systems are awaiting stars. Additional research tracks — governance runtimes, change intelligence, and workflow primitives for autonomous software — are uncharted for now.
- 🌐 Company site — https://horonom.com
- 🤖 AI Agent Assembly org — https://github.com/ai-agent-assembly
- 🤝 Contributing
- 🔒 Security policy — report privately to
security@horonomy.dev