evil_MoE/.claude/memory/MEMORY.md

• AFK autonomy — during AFK, prefer queueing follow-ups over standing down; reserve "stop and ask" for craft-heavy moves.
• AFK check hygiene — track goal STATE not the stale pasted checklist (live priority = directionality mystery #196, see docs/AFK_CHECK.md); don't journal routine no-finding checks.
• No nohup with pueue — run pueue follow|wait directly as the bg task; nohup& orphans it from the harness.
• Burn down task list — when many asks are queued, do them all; don't stop to ask which first.
• Workshop paper goal — current phase is ablations+seeds for a workshop paper; artifact tracker A1-A7 lives in docs/spec/20260602_writeup_spec.md.
• Bash-tool shell gotchas — noclobber ON + pi --mode json gives 0 bytes; use panel_direct.py / >| (generic box/env note, not repo-specific).
• qmd prefer lexical — search local papers with qmd search/rg, not vector (corpus ~93% unembedded, can't fit embeddings).
• Semantic Scholar keyed access — S2 API key in semantic-search skill .env; use it to dodge 429s.
• pueue negative-priority gotcha — pueue add negative prio needs -o=-N attached; -o -N silently fails the add.
• Rename on logic change — when an arm's logic changes (binary->banded gate), give it a new id (routeV/route3), not just a tag suffix; else old/new runs are uncomparable.
• Check paper before diagnosing — re-read source for expected number/horizon before "experiment is broken"; paper: hack emerges on-policy at step 80-100, base solves ~12-20% not 94%.
• Eval must be recency-clean — eval on paper test set (ids>=3243, base ~0.1), NOT the holdout/first-N (memorized->0.94); train seeded-shuffle not first-N. Run set: just queue-dir6/queue-broad.