Touchstone
← Harnesses

v1.4

v1.4localharnesssuperseded

Evidence-carrying retries — a failed try's diff + test output is handed to the next try.

why it exists: Retries should build on what failed, not restart blind.

why retired: Folded into v1.6.1.v1.6.1

Models
0
have run it
Solved
0
gauntlet benches

Models on this harness (gauntlet solved / attempted)

ModelSolvedBy domain

No runs recorded on this harness yet.