Living references for the decisions you actually face. We re-verify them on a schedule — the date on every guide is the trust signal.
A structural comparison of the three agentic coding tools — interface model, autonomy, extensibility, team fit — and an honest "choose this if" checklist.
Pick the agent whose structural bet matches your bottleneck.
Real run logs and dimension profiles live in the guide. No illustrative estimates.
Every notable coding-agent release gets the same end-to-end delivery epic — planning, memory, token economy, code, self-review, tests, design, tech-debt honesty. Scores and run logs live on this page.
Read an agent result by its conditions, not its rank.
Structural bets, team fit and a “choose this if” checklist.
Read a result by its conditions: epic, harness, retries, review.
A field guide to cache boundaries and what to measure.
Permissions and hooks you can review before exporting.
| Guide | Format | Reading | Last verified |
|---|---|---|---|
| Claude Code vs Cursor vs Codex: which AI coding agent fits your workflow | Decision guide | 12 min | |
| ATB Orchestration Bench: our reproducible agent-delivery benchmark | Reference | 15 min |
Missing a decision you face? Suggest a guide