Snapshots

Nothing a model invents reaches a shopper without a guard looking first

Settled
  • Five kinds of claim are flagged: codes, guarantees, timed results, absolutes and efficacy.
  • Every picture gets a product check and a look judge; a second failure is held back.
  • Fact gate, Play check and proof sheet keep every number tied to a source.
  • Guard tests run on every deploy, and a failing suite stops it.
  • Nobody counts how often each guard fires yet.
47of 54 pictures were fine
Your call

Log one daily line per brand with flags, mends, redos and holds?

AI and models · 6.3

Guards

DraftLast checked Mon Oct 26Owner Param

Answer

The checks that sit between a model's output and a person: what each one catches, what it does when it fires, and the rule that every guard works for every brand and heals saved work, so nothing a model invents reaches a shopper unchecked.

Today

Words: claims

claimFlags(text, source) flags five kinds of claim the brand's own words do not carry brew/claims.js:64-82.

Kind Example it catches Passes when Source
code "Use code PURE25" The code is on the store brew/claims.js:24-25
guarantee "money back", "risk free" The store says it brew/claims.js:26
timed "visible results in 14 days", "overnight results" The store says the same time brew/claims.js:28
absolute "clinically proven", "dermat-approved", "100% natural", "cures" Health words need the study, test or certificate beside them in the store's words, or a Brand Book approved claim; other absolutes need the store's own words brew/claims.js:32 brew/claims.js:74-78
efficacy "up to 60% less hair fall", "reduce hair fall by 7.4X" The store names where the number was measured, or the claim is approved brew/claims.js:36-37
  • "Dermat-approved" passed once on Arata and is now caught in every spelling brew/claims.js:30-31.
  • An approved claim counts only inside the Brand Book's approved list, never anywhere in the brand's words brew/claims.js:50-52.
  • Every writer prompt carries CLAIMS_RULE brew/claims.js:88.

What happens when a claim fires depends on the surface:

Surface Action Source
A Play Sentence kept, marked "Proposed, team decides"; a claim in a title or cover number goes to To confirm brew/claims.js:102-137
A draft made with nobody watching Every flagged sentence dropped; a flagged headline never set on the picture brew/claims.js:143-153
A feed card that no person rewrites Sentences with codes or guarantees dropped brew/claims.js:93-97
Anything a shopper, team or partner may copy checkCopy drops flagged sentences, never throws brew/claims.js:215-232
A film's spoken line Claim taken out before the model hears it brew/jam2/studio.js:133
A still's words A claim, or a price that is not this product's own, is taken out brew/jam2/studio.js:134-142

Words: banned words, copy and slips

  • bannedHits finds category words ads cannot use, but never inside quotes, a denial, an avoid list, or a word the brand itself uses brew/claims.js:110-118. The banned list is config plus the Brand Book never-say list memory brew-brand-book.
  • copyFlags flags shouting caps, forced slang, nonsense time phrases, repeated words and literal "\n" brew/claims.js:166-195. Every writer carries COPY_RULE brew/claims.js:196.
  • writerSlips adds the "isn't A; it's B" turn, paise on a price the store shows whole, and for WhatsApp or SMS: dash lists, label lines, over 700 characters brew/claims.js:203-213.
  • Mend: a jam package with slips is rewritten once on exactly those slips before anyone sees it, so "Needs a fix" never lands on a fresh draft brew/remix.js:642-650. Added 26 Oct after an Arata WhatsApp broadcast brew/claims.js:199-202.
  • Every structured reply gets real line breaks, and ₹ written as ‹ is fixed brew/ai.js:9-11.

Numbers and proof

  • Fact gate: every money amount and percentage on a card must be on the fact sheet; strict mode also gates multipliers and large counts brew/fact-gate.js:1-12.
  • Honest: a story quotes only numbers in its data, small counts up to 10 aside brew/research.js:57.
  • Play check: each Play is read by a second model against the lines it cites, and only flagged sentences are fixed; an unbacked number becomes a labelled target brew/moves-engine.js:553 brew/moves-engine.js:568.
  • Proof: the proof sheet must prove the claim it sits under; prices, plan rows and Plays are never proof brew/proof-logic.js:128 memory brew-proof-plain-make.
  • Brand Book verify: a line without a citation or a verbatim quote is dropped brew/brand-book-read.js:205. The Brand Book has its own audit and guard test before every deploy memory brew-brand-book-guards.

Pictures: the image guard

One wrapper around the image model gives every caller the same checks brew/image-guard.js:1-24.

Step What it does Source
Reference The product the words name, matched to the catalog; its plainest photo; never a logo brew/image-guard.js:74-90
Keep The prompt says to keep the product exactly as photographed brew/image-guard.js:29
Product check SAME keeps it; TEXT remakes once on the other model, turned and smaller; DIFFERENT remakes stricter, then the real photo stands in; NOREF tries the next catalog photo brew/image-guard.js:110-140 brew/server.js:212-230
Look judge A free pixel check, then a vision judge: GOOD, COLLAGE, CROPPED, OFFTOPIC, EXPERT brew/image-guard.js:57-59 brew/draft-quality.js:91-98
Redo or hold back A failure is made once more with the reason; a second failure is held back and never shown brew/image-guard.js:60-72
Retry Up to 3 tries with backoff; a blocked or refused prompt is not retried brew/image-guard.js:43-50
  • The pixel check flags two flat grounds of different colours at the frame, or a blank band; it is strict on purpose, FLAT 2.5 and 10% of the frame brew/draft-quality.js:37 brew/draft-quality.js:56-68 memory brew-draft-quality.
  • EXPERT fails any person shown as a doctor or scientist, a fake endorsement; saved drafts were judged again at LOOKS_V 2 brew/draft-quality.js:15 brew/draft-quality.js:97.
  • When in doubt the judge says GOOD; split screens pass and page mocks are never judged brew/draft-quality.js:93 memory brew-draft-quality.
  • See-through packshots are flattened on white before the model sees them memory brew-draft-quality.
  • A cut-out with no real matte or a jagged edge is never placed brew/draft-quality.js:73-88.
  • Of 54 pictures the old check threw away on 16 Oct, 47 showed the right product, so blurred small print now counts as SAME brew/server.js:218.

Studio and finals

  • Critic: each Studio shot is judged against the product photo; a check that did not run never buys a second picture brew/jam2/creative.js:99-109.
  • Made check: every word a designed piece drew is read back; an invented size, number, claim or fake review fails it. Higgsfield once printed "10ml" on a 30 ml bottle and set a tagline as a 5-star review brew/jam2/creative.js:111-131.
  • A take that fails its checks is retried once, never more brew/jam2/studio.js:1-5.
  • A testimonial style is blocked when the Brand Book says no testimonials brew/jam2/studio.js:88.
  • Video spelling: after a video final lands, two frames per worded beat are read and compared with the words asked for; nothing is remade by itself because a remake costs credits brew/text-check.js:1-5 brew/text-check.js:50-55. Added after Arata's reel read "concrate" memory brew-viral-remakes.

Tests that hold the guards

Guard tests run on every deploy with AI_MOCK=1, and a failing suite stops the deploy brew/deploy.sh:12-16. Files: claims, claim-kinds, claim-rules, round2-claims, writer-slips, make-claim, fact-truth, honest-copy, proof-logic, image-guard, draft-quality, text-check, brand-book-guards, qv3-guards brew/test/.

Tomorrow

Item Version Tag
Every guard brand-agnostic, healing saved work on read v1 Stays
A labelled claim corpus that measures precision and recall of claimFlags v1 New, see Evals
Guard verdicts logged per brand and per day, so a rising hold-back rate shows v1 New
Video spelling check that offers one remake with the price shown v2 Changes
A send gate: nothing posts or schedules without guards passing and a person's yes v2 New
Ad policy checks per platform before an ad runs v4 New
Category rule packs, such as ASCI, FSSAI, FTC, ASA, kept as Brand Book suggestions until the team keeps them v1 Stays memory brew-brand-book

How we do it

  1. A new surface that a shopper, team or partner can copy calls checkCopy or checkCopyIn with packSource(pack, card) brew/claims.js:57 brew/claims.js:215-232.
  2. A new picture caller goes through the image guard, never the raw model.
  3. A new incident class gets a guard in code, a test, and a heal for saved data, the same day.
  4. Run the guards locally:
    cd /Users/param/Documents/ChatGPT/brew
AI_MOCK=1 node --test test/claims.test.js test/writer-slips.test.js test/image-guard.test.js test/draft-quality.test.js test/text-check.test.js
  5. Turn the look judge off only to diagnose: LOOKS_CHECK=0. IMAGE_CHECK=0 turns off both picture checks brew/server.js:232.
  6. Before a Brand Book deploy: node tools/audit-brand-book.mjs DIR and the guards test memory brew-brand-book-guards.

Numbers

Metric Definition Where measured Status
Hold-back rate Pictures held back after two look failures, over pictures made guard.held on each picture brew/image-guard.js:70-71 Proposed, not counted today
Wrong product shipped Pictures shown whose product is not the catalog product spot check Proposed, target 0
Mend rate Jam packages rewritten on slips, over packages made remix.js mend Proposed
Claim flags per 100 drafts Flags removed by Clean cleanDraftClaims return Proposed
Video spelling pass Finals whose on-screen words match text-check result Proposed

Open

  1. Count the guards. Nothing records how often each guard fires. Proposal: one daily line per brand in the ops page with flags, mends, redos and holds.
  2. Judge model. The picture checks ask for Haiku and run on Flash today. Proposal: keep Flash until the eval set shows judge agreement with a person below 90%, then move the judge to Claude.
  3. Video spelling remake. Proposal: keep it manual in v1, and in v2 offer one priced remake.
Sources
  • brew origin/main b91e364: claims.js:1-232, image-guard.js:1-142, draft-quality.js:1-109, text-check.js:1-55, fact-gate.js:1-12, research.js:57, proof-logic.js:128
  • brew origin/main: server.js:212-236, remix.js:622-650, moves-engine.js:540-575, jam2/creative.js:95-135, jam2/studio.js:88 and 133-142, brand-book-read.js:205, ai.js:9-11
  • brew origin/main test/: claims, claim-kinds, claim-rules, round2-claims, writer-slips, image-guard, draft-quality, text-check, fact-truth, honest-copy, proof-logic, make-claim, brand-book-guards, qv3-guards
  • memory brew-draft-quality, brew-brand-book-guards, brew-proof-plain-make, brew-sloosh-finals, brew-viral-remakes, brew-jam-v2, brew-setup-v2, brew-qa-audit-2026-10-20

Source file content/ai-and-models/guards.md · For agents: Markdown · JSON