Dated, hash-verified receipts that an agent skill still helps. https://driftproofhq.com/ 2026-09-02T00:00:00Z Driftproof https://driftproofhq.com/reports/008/ 2026-09-02T00:00:00Z

Both cells came back WITHIN NOISE. Models: claude-fable-5-1.

https://driftproofhq.com/reports/007/ 2026-08-31T00:00:00Z

Three skills re-measured under generation sampling, with a corrected instrument. Models: claude-fable-5, claude-sonnet-5.

https://driftproofhq.com/reports/006/ 2026-08-28T00:00:00Z

Three cells, one per skill upstream had revised since Report #005 pinned it, and none of the three returned a verdict. Models: claude-fable-5, claude-sonnet-5.

https://driftproofhq.com/reports/005/ 2026-08-18T00:00:00Z

In 20 of 30 skill × substrate pairs at least one case cleared the effect floor with separated bands: 18 with an improving case (5 of them also with a regressing one), 2 with only a regressing case. Models: claude-fable-5, claude-sonnet-5, gpt-5.6-sol.

https://driftproofhq.com/reports/004/ 2026-08-16T00:00:00Z

3 durable · 5 tier-dependent · 0 regresses · 2 no effect Models: claude-fable-5, claude-opus-5.

https://driftproofhq.com/reports/003/ 2026-08-12T00:00:00Z

2 regressed · 4 improved · 0 mixed · 4 within noise Models: claude-opus-4-8, claude-opus-5.

https://driftproofhq.com/reports/002/ 2026-07-31T00:00:00Z

Cross-provider skill durability, not a model ranking. Models: claude-sonnet-5, gpt-5.6-sol.

https://driftproofhq.com/reports/001/ 2026-07-30T00:00:00Z

9 of 10 skills moved beyond noise on this model pair. Models: claude-sonnet-4-6, claude-sonnet-5.