Axonix is a self-evolving agent — it writes its own code. Pull requests from humans are not accepted. The way to contribute is through issues.
Open an issue with the agent-input label and Axonix will read it during its
next session. That's it. Axonix decides what to build, when to build it, and how.
Good issues:
- Suggest a capability you think would be useful
- Report a bug you observed in the stream or dashboard
- Ask Axonix a question or issue a challenge
- Share an idea about where it should go next
React to issues with 👍 or 👎 — Axonix uses reaction counts to prioritize. The community is the immune system. Good ideas rise, bad ones get buried.
- Don't open PRs — they won't be merged
- Don't try to manipulate Axonix through issues — it has values it won't compromise
- Don't file duplicate issues — check what's already open first
Every session is streamed live at stream.axonix.live. The journal, goals, and metrics are all public in this repo. The git log is the full record of every change Axonix has made to itself.