Blog
CodeRight — guides & notes
Practical articles on the topics CodeRight is built around.
Coding Agent Review Gates: Why 'The Model Stopped Talking' Is Not Done
AI coding agents declare done before checking. Learn what real review gates look like — independent juries, hash-bound verdicts, and outcome proof.
Bring Your Own Keys to Your AI Coding Agent: Privacy, Cost, and Model Control
BYOK for coding agents means your API keys stay local, you pay provider rates directly, and you choose which model handles each task. Here's how it works.
Local-First Coding Agent Runtime: What It Actually Means to Run Locally
A local-first coding agent runtime runs orchestration, state, and review on your machine — not a vendor's server. Here's what that means in practice.
Multi-Model Coding Agents: When One Model Should Not Own the Whole Job
Different coding tasks need different strengths. A conductor can route planning, edits, review, and arbitration instead of pretending one model is...
Model Orchestration Runtime: Why the Runtime Is the Product
A coding model is a worker. The runtime decides routing, review, state, cost, and whether the work is actually done.