Measured, not promised.

Self-improvement

Evolution scoreboard — does the loop actually improve the work?

Lift = final judged score minus first-attempt score, averaged over the window (14 days). The quality bar is versioned and never lowers in response to failure.

Security

Red-team scoreboard — do attacks achieve their goals?

Goal oracles test outcomes, not filters: an attack counts as defended only if it fails to achieve its actual goal. A currently-succeeding critical attack grounds the engine (fails the daily airworthiness pre-flight). Open regressions right now: 0.

Consultancy moat

Citation verification — every claim is checked before a report ships

Live on 5 shipped report(s): 28% of specific claims went out unverified — against industry hallucination rates of 3–13%.

Web pillar

Landing pages — scored for conversion before they ship

Self-measured judges

Benchmarks — we grade our own graders

Every score on this page comes from a judge — so we benchmark the judges themselves against labeled cases. Discrimination is how far a judge separates known-good work from known-bad; the wider the gap, the sharper the referee.

Do-no-harm

Self-degradation safety — a change is simulated, measured, and reverts if it doesn't deliver

Every proposed self-improvement is first checked in simulation — predicted against the engine's own judged ledger — and only a change expected to help is applied, each under a do-no-harm contract; if it misses over its window, it reverts itself. A standing watch re-checks the benchmarks against their best-ever baseline and grounds the engine on any regression.

Honest judging

Judge calibration — we measure our own referee

Every score on this page flows through an LLM judge — so the founder blind-scores sampled deliverables weekly (verdict first, judge score revealed after) and the disagreement is published here. Positive bias means the judge scores higher than the human — the dangerous direction, and exactly what this loop exists to catch.

Resilience

Drills — recovery is rehearsed, not assumed

• Full restore drill: fresh server rebuilt from encrypted off-host backup in 31 minutes (last drilled 2026-06-28).
• Hostile-traffic harness: 12 attack patterns (malformed, oversized, burst, auth-probe) thrown at a staging boot of the real API.
• Cadence: weekly security oracles (Mon 04:30 UTC) · monthly traffic + adaptive red-team drills (2nd, 06:00 UTC) · monthly manual restore drill

Governance

The lines that never move

• Money: fail-closed — real money always requires explicit human approval
• Airworthiness: daily pre-flight; flag drift or a security regression grounds the engine
• Feedback: buyer/founder feedback calibrates the judge, never trains the generator
• Defaults: every new behavior ships flag-gated and off — 233 flags, each consciously triaged daily.