0xrsydn/sts2-bot
1
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

docs(api): record reward observation audit and deployment status

Browse source
This commit is contained in:
0xrsydn 2026-09-23 18:25:36 +07:00
commit d826795bbb
2 changed files with 132 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

2
docs/COMPOSITE_REWARDS.md
View file

@ -68,3 +68,5 @@ selection, and recording. No live game or model connection is needed.
See [the research proposal](research/15-composite-card-policy-proposal.md) for the
full algorithm and evaluation plan. This implementation is its first, limited stage.
The [reward API audit](research/16-reward-api-audit.md) identifies the observation
changes needed before completing take-versus-skip decisions.