CommunityJev Integrations & Adapters
Brand Safety Feed
A contextual brand-suitability scorer that judges each feed item against an advertiser's own versioned policy, returning a category, an exclusion flag, and a confidence for routing borderline items to review.
- Published by
- Community
- Author
- Adguard Labs
- Last verified
- Oct 2026
- Languages
- TypeScript
- Capabilities
- Open sourceRequires API key
Highlights
- Scores against an advertiser's own versioned policy instead of one universal harm taxonomy.
- Logs the policy version with every decision so a later review can explain an exclusion.
- Keeps keyword and domain blocklists as the first deterministic pass, catching context after.
- Batches items as parallel questions so the running cost tracks input tokens.
Watch out
It is a contextual layer that complements, not replaces, blocklists and human review; a policy edit should be evaluated on the bundled labelled feed before it reaches live inventory.
Questions about this entry
Short answers drawn from the linked project or page.
It scores each item against a brand's own suitability policy rather than a universal harm taxonomy, so the same post can be brand-safe for one advertiser and not another, and the policy is data the operator edits.
A Choice over the policy categories, a Noul flag for outright exclusion, and a confidence the feed uses to route borderline items to a human queue instead of blocking them.
Yes. It batches items into parallel questions within the 64k context and reports a steady-state throughput on the sample feed, so the cost is input tokens rather than output.
No. It complements keyword and domain blocklists by catching the contextual cases they miss, and the README keeps the blocklist as the first, deterministic pass.
Policies are versioned YAML files checked into the repo, and every decision logs the policy version so a review can explain why an item was excluded after the rules changed.
Node 22 or newer and a Typesafe key. A sample feed of 5,000 labelled posts ships with the repo for evaluating a policy edit before it reaches production inventory.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
YouTube: Building a contextual brand-safety scorer on Jev from scratch.
Hacker News: Show HN: Ad Spend Guard gates budget before the auction, not the creative after.
Show HN: Gating ad budget with a decision modelRead the thread on Hacker NewsX: A website slop detector that checks 35 tells in 243 ms
X: Breaking, golden nugget, or slop
More Jev resources like this
A pre-bid spend guard that scores a placement's expected relevance and flags whether it matches the campaign brief, letting deterministic code hold budget back before the auction instead of gating the creative itself.
Jev Integrations & Adapters#community#typescript#advertising
Community
1GitHub stars
Cloudflare Worker that asks Jev on Workers AI for literal profanity in text or usernames and, for usernames, a second Noul for phonetic or look-alike disguises. The threshold and max policy live in Worker code, with OpenAPI docs generated via chanfana and Hono.
Jev Integrations & Adapters#community#cloudflare-workers#moderation
Community
4rays/profanity-checkerTypeScript
4rays/profanity-checkerTypeScript12.4kGitHub stars
Open-source AI email assistant (Next.js, Postgres, 12k+ stars) with a decision-model layer that can run on openrouter:typesafe/jev-1.13: Jev-backed questions cover rule selection, cold-email detection, thread status, recurring patterns, unsubscribe pages, and sender categorization, each with evals.
Jev Integrations & Adapters#community#email#gmail
Community
elie222/inbox-zeroTypeScript
elie222/inbox-zeroTypeScript689GitHub stars
The Vercel AI SDK provider for OpenRouter, including openrouter.decisionModel for typesafe/jev-1.13 evaluations with providerMetadata answers.
Jev Integrations & Adapters#community#typescript#openrouter
2GitHub stars
MIT-licensed Chrome extension that asks Jev whether the page, video or post you are on serves your written goals -- a category Choice, a goals Noul and a 0-4 waste-of-time Score -- then blocks or blurs in code, tightening thresholds at 2, 10 and 20 minutes of engaged time.
Jev Integrations & Adapters#community#chrome-extension#digital-wellbeing
Community
plusminushalf/rot-guardJavaScript
plusminushalf/rot-guardJavaScriptA real-estate matching tool that applies hard-constraint gates before ranking a buyer's property preferences against listing features and remarks, returning a fit band and a per-feature score for the agent.
Jev Integrations & Adapters#community#typescript#real-estate
Community
From the Jev community
Posts from builders shipping with Jev right now.