Evaluation receipt / agent-receipt

Agent Receipt

Both arms passed the corrected evidence-binding checks, and all three blind comparisons tied.

Inconclusive
0/3Treatment wins
3/3Treatment verifier passes
3/3Baseline verifier passes
0Tool errors

What this proves

Produced public-safe receipts without demonstrating a measurable advantage.

Inspect packageMachine record
Status

inconclusive. The corrected comparison demonstrated no lift over the empty baseline.

Method

Three blind paired Harbor runs against the same agent without the skill.

Model

openai/gpt-5.6-sol

Limitations
  • One representative task
  • One model
  • Three valid pairs
  • Earlier undisclosed-schema comparison was invalidated
Exact packagesha256:fa93aa2dc0b9b4bc32c421ef972762c3f62c968a37a1b5497b73dfa81d2e0139