yolo: log backlog + mid-run status (8 shipped, continuing)

0c1d69154078e3f469cf9c7b1da2f9a41d118d57 · 2026-07-22 23:52:23 -0700 · Steve

Files touched

M yolo/IDEAS.md

Diff

commit 0c1d69154078e3f469cf9c7b1da2f9a41d118d57
Author: Steve <steve@designerwallcoverings.com>
Date:   Wed Jul 22 23:52:23 2026 -0700

    yolo: log backlog + mid-run status (8 shipped, continuing)
---
 yolo/IDEAS.md | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/yolo/IDEAS.md b/yolo/IDEAS.md
index 27228f5..52c6994 100644
--- a/yolo/IDEAS.md
+++ b/yolo/IDEAS.md
@@ -21,3 +21,13 @@ Answering the /contrarian critique ("one builder vote = theater", "toy demos not
 - [DONE] #6 Daily auto-challenge generator — scripts/daily-challenge.js creates a $0 all-local battle on a day-rotating prompt, renders + AI-judges, logs to yolo/daily-log.jsonl (no auto-crown, keeps human vote pure). Test-run OK. launchd install DRAFTED to pending-approval (GATED, not installed).
 - [DONE] #7 Remix mode — feed a battle's best artifact's ACTUAL HTML (capped 14KB) back to the models as 'improve on this', spawning an evolutionary next-round battle (remixOf/remixSource lineage). Verified: remixed the product page, gemma produced an improved version. $0 local default.
 - [DONE] #8 Source view — </> per pane swaps the render for the raw HTML (scrollable), toggle back. README updated with all 8 features.
+
+## Backlog for next /yolo iterations (this run continues ~2h)
+- [ ] Multi-vision-model CONSENSUS referee — have all 4 local vision models (qwen2.5vl, minicpm-v, llava, moondream) score each artifact; robust consensus pick + variance.
+- [ ] Tournament bracket mode — single-elim across N models, auto-advance winners.
+- [ ] Per-category leaderboards — tag challenges (Games/Real Work) + per-category ELO ("who's best at real work?").
+- [ ] Export/share a battle as a standalone static HTML page.
+- [ ] Model profile page — all of a model's artifacts + its stats over time.
+- [ ] Head-to-head 1v1 quick mode.
+
+## Status @ ~27min: 8 features shipped + committed, 7/7 six-way clean, 1 gated draft (daily launchd). Live: modelarena.agentabrams.com

← 0688aaa yolo idea 8: artifact source view (</> per pane) + README do · back to Model Arena · yolo idea 9: multi-vision CONSENSUS referee — panel of local d2bb069 →