Commit Graph

2 Commits

Author SHA1 Message Date
05b72f6987 fix: ignore markdown emphasis when matching read answers
Some checks failed
CI / test (pull_request) Failing after 51s
The read checker matched a task's required-fact phrasings as plain
substrings of the agent's report after only lowercasing and collapsing
whitespace. An answer that was substantively correct but wrapped a value in
markdown (e.g. `**5**`) failed the match, because the emphasis markers broke
the phrase adjacency (`**5** open` does not contain `5 open`) — a correct
answer scored incorrect on formatting alone.

Strip markdown emphasis/code markers (`*`, `_`, backtick) during
normalization so the match is on substance, not presentation. A guard test
confirms a wrong value still fails after stripping.
2026-07-17 11:55:53 -04:00
66f576a2b7 feat: add benchmark checker and scoring spec (task 0024)
All checks were successful
CI / test (pull_request) Successful in 51s
CI / test (push) Successful in 48s
2026-07-16 07:33:31 -04:00