driftproof run — skill "documentation-and-adrs" v0.0.0
  content_hash: b67a9f07ed105796…  suite: 1 cases (6cce54e7d9d0…)
  surface: claude-cli   models: claude-sonnet-5-5   samples/case: 3   concurrency: 1
  registry: claude-sonnet-5-5:registered   transcripts: retained-local
  judge: claude-opus-5   (--judge-model)
  isolation: same-user (--trusted-skill: self-authored skill)
  projected calls: 80/model × 1 model(s) = 80   per-model cap: 80
  projected cost: ~$0.94 (rough upper bound; budget $8.00, hard-stop $10.00)
  actual metered spend on claude-cli: $0.00 (subscription; the $ figure is the estimated-equivalent API cost, counted against the cap identically)

── model: claude-sonnet-5-5 ──
  comment-intent-not-implementation / with_skill: borderline (0.78 ± 0.14)
  comment-intent-not-implementation / baseline: fail (0.57 ± 0.03)
  → transcripts (retained-local): transcripts/47fc7a4455f21f1f848f3b4968cfbc744ed5791cf363f1c64a4ab9228444b20c/
  → ../runs/r013-20260929T035323Z/claude-sonnet-5-5/documentation-and-adrs/receipts/documentation-and-adrs-claude-sonnet-5-5-2026-09-29.json  (36 calls)
  → answered by: model claude-sonnet-5-5 (attested by the surface; the same-user spawn (--trusted-skill)); verification_level TESTED
  → skill lift +0.208 ± n/a (1 case)  (with 0.781 ± n/a (1 case) vs base 0.572 ± n/a (1 case); band = sample stddev of the per-case means)

Done. 1 receipt(s) emitted to ../runs/r013-20260929T035323Z/claude-sonnet-5-5/documentation-and-adrs/receipts/