# FSOT Credibility Hardening Audit

*Generated: 2026-07-16T15:52:59.069796+00:00*

Hard credibility expansion — every pillar must be independently reproducible, not rhetorical.

**Score: 14/14 pillars green**

| Pillar | Status | Evidence |
|--------|:------:|----------|
| Formal triangulation | ✓ | Lean+Coq+Isabelle+F*+Rust |
| Benchmark green gate | ✓ | 405/405 |
| Zero free parameters | ✓ | ZERO_FREE |
| Contested sectors | ✓ | H₀/σ₈/BBN panel |
| Near-miss published | ✓ | transparent worst greens |
| Preregistration locked | ✓ | PRED-001–041 |
| Wet-lab longevity | ✓ | Tier 94/95 |
| Live catalog ingest | ✓ | Gaia/GWOSC/NEO refresh |
| Genetics crosswalk | ✓ | Tier94↔Tier95 |
| Skeptic kit | ✓ | 15-min path |
| Lean route depth | ✓ | 7/7 routes |
| Circuit Phase 1 | ✓ | Tier 96 BOM emergence |
| Live ingest schedule | ✓ | weekly refresh policy |
| Practical pipeline | ✓ | local application + blueprints |

## Headline metrics (live)

- Benchmark green: **394/394**
- Pooled median: **0.013003%**
- Atomic obligations: **1863**
- Contested pooled: **0.029748999999999998%**

## Deferred (convenience, not math)

- **ESP32 hardware observer** — eight-way hardware closure deferred until boot-sequence workflow is ergonomic (laptop bench setup). Formal spine and QEMU bare-metal remain authoritative.

## Regenerate

```bash
python scripts/run_publication_verification_bundle.py
python scripts/build_wetlab_longevity_expansion_bundle.py
python scripts/build_lean_route_credibility_expansion.py
python scripts/build_circuit_component_expansion_bundle.py
python scripts/build_credibility_depth_bundle.py
python scripts/build_credibility_hardening_audit.py
```