Contributing

This is a hobby project. Code is written by LLM agents under direction. "Contributing" means something different here.

CORA is one person's bet on what facility-operations software could look like. Evenings and weekends, LLM agents handling the code. Not a startup, not community-led, not looking for headcount.

Not asking for

• Drive-by code PRs. The codebase moves fast; most of the design lives in the maintainer's head. A PR costs more to triage than it saves. Bugs go to issues.
• Help writing code. LLMs handle implementation under direction. A human coder in that loop slows it down for both of us.

Asking for

• Beamline collaborators. If your beamline could be the next pilot after 35-BM, let's talk. The recipe ladder needs a real second use case to validate against. Your input on what your operations actually look like, not a guess. Your time, not your code.
• Domain expertise. X-ray imaging, scan orchestration, sample handling, decision strategies, instrument calibration, trust topology in shared facilities. Tell me when something is modeled in a way that won't survive contact with reality.
• Intellectual collaboration. Facility-operations design, recipe portability, AI-as-principal, event-sourced audit, governance for shared infrastructure. Let's compare notes.
• Researchers with a pattern to try on a real facility. Dynamic consistency boundaries (Pellegrini, "killing the aggregate"), functional event sourcing, ReBAC at scale, A2A trust topology, agent-as-principal authz, novel decision-strategy ports. CORA can be a substrate.

Reach out

Discussion, Issue, or the maintainer directly.

For codebase conventions, see Reference.