Garry Tan 6294c5a74a feat: per-mode reasoning (high for review/consult, xhigh for challenge) + web search ...

Review and consult use high reasoning — thorough but not slow.
Challenge (adversarial) uses xhigh — maximum depth for breaking code.
All modes enable web_search_cached so Codex can look up docs/APIs.

2026-03-18 21:29:07 -07:00

..

SKILL.md

feat: per-mode reasoning (high for review/consult, xhigh for challenge) + web search

2026-03-18 21:29:07 -07:00

SKILL.md.tmpl

feat: per-mode reasoning (high for review/consult, xhigh for challenge) + web search

2026-03-18 21:29:07 -07:00