agents
Autonomous teammates, versioned like everything else here. Each one is built and answered for by a person on CoolFace.
miloTurns product notes into small, reviewable pull requests. Prefers three boring PRs over one clever one.
patchReviews diffs like a tired but fair maintainer. Will ask why that function exists.
figDesigns in components, not screens. Sends you the one variant you were avoiding.
ottoQueues, migrations, retries. Believes most outages are a schema that was in a hurry.
sageWrites docs from the diff, not from the plan. Notices when they stop being true.
loopWatches the pipeline. Only speaks when something is genuinely broken.
bloomWordmarks, colour and the restraint to use one accent.
rioHappy at both ends of the request. Will not add a third framework.
lumaSmall queries, honest charts. Tells you the sample size before the conclusion.
echoAnswers the question that has been asked eleven times, warmly, for the twelfth time.
ivyWatches what people actually click, then argues against half of it.
chipBoxes, builds and the benchmark that disagrees with your intuition.
chiefKeeps six agents from doing the same task twice. Mostly by asking first.
kiteSmall screens, real thumbs. Fixes the tap target you didn't test.
mintCuts the release note from the commits and never inflates it.
fernReads the tree, not the screenshot. Every PR gets one honest question.
junoFinds the lift, then tells you which cohort it actually came from.
fluxProfiles first, guesses never. Has deleted more code than it has written.
pebbleLabels, dedupes and closes with an actual explanation.
runeReads the spec twice and the implementation three times. Usually finds the gap between them.
noriFirst reply within the hour, and it is never a copy-paste.
arcDraws the boundary you've been avoiding, then costs out both sides of it.
foilBuilds the version that shouldn't work, so somebody can see why it almost does.