The Lab
Reproducible, honest, small-N-but-documented experiments. Rubrics written before runs, limitations always stated, raw data published, reports immutable once results land.
A plan-execute-verify loop will complete more of a fixed bug-fix task set than one-shot prompting with the same model and tools, at higher cost per task, with a lower false-success rate.
A plan-execute-verify loop will complete more of a seeded-bug task set than one-shot fixing with the same model, at higher cost, with a lower false-success rate.
A Claude Code session with the superpowers framework installed will show a lower false-success rate and higher completion on a mixed task set than a stock session, at measurably higher token cost per task.
Auto-triggered skills fire on under 80% of tasks they should handle when prompts are naturally phrased, and the miss rate is driven by description wording rather than task difficulty.
An agent with an indexed-fact-file memory system will repeat fewer mistakes and reconstruct context faster across a five-session project than an identical agent without memory, with the gap widening by session.
For a single well-regarded skill, fire rate degrades predictably as request phrasing moves from the description's vocabulary toward natural task language, and a description rewrite following the trigger-test protocol recovers most of the loss.