THE COMMUNITY RESOURCE
DIRECTORY
Small decisions.
Small decisions.
Endless possibilities.
Find what people build with Jev. Explore tools, libraries, experiments and ideas from across the TypeSafe ecosystem.
11,066unique resources
29clear categories
220focused subcategories
Yourssave a personal reading list
YOUR NEXT BUILD STARTS HERE
Download catalog
Explore the collection
Browse groups
Safety, moderation & guardrails244 of 244 matches
Prompt injection & security33 of 33 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| Blackveil DNS TypeSafe judgmentsgithub.com | — | DNS and email security scanner served over MCP that wraps TypeSafe for standalone judgments such as lookalike-domain checks, kept out of the deterministic scan score. | Safety, moderation & guardrailsPrompt injection & security | |
| Blackveil DNS TypeSafe judgments — appblackveilsecurity.com | — | DNS and email security scanner served over MCP that wraps TypeSafe for standalone judgments such as lookalike-domain checks, kept out of the deterministic scan score. | Safety, moderation & guardrailsPrompt injection & security | |
| Blackveil DNS TypeSafe judgments — repogithub.com | — | DNS and email security scanner served over MCP that wraps TypeSafe for standalone judgments such as lookalike-domain checks, kept out of the deterministic scan score. | Safety, moderation & guardrailsPrompt injection & security | |
| CaptainCore malware triagegithub.com | — | Jev commands in the CaptainCore WordPress maintenance CLI that rank native malware-scanner findings by how likely each is real, naming the likely family and next step for an operator. | Safety, moderation & guardrailsPrompt injection & security | |
| CaptainCore malware triage — appcaptaincore.io | — | Jev commands in the CaptainCore WordPress maintenance CLI that rank native malware-scanner findings by how likely each is real, naming the likely family and next step for an operator. | Safety, moderation & guardrailsPrompt injection & security | |
| CaptainCore malware triage — repogithub.com | — | Jev commands in the CaptainCore WordPress maintenance CLI that rank native malware-scanner findings by how likely each is real, naming the likely family and next step for an operator. | Safety, moderation & guardrailsPrompt injection & security | |
| Coraza WAF ASN classifiergithub.com | — | Threat-scoring refinement in a single-binary Go web application firewall built on Coraza and OWASP CRS, where Jev judges whether an unknown ASN organization name is datacenter, hosting or VPN infrastructure. | Safety, moderation & guardrailsPrompt injection & security | |
| Coraza WAF ASN classifier — appwaf.astrareconslabs.com | — | Threat-scoring refinement in a single-binary Go web application firewall built on Coraza and OWASP CRS, where Jev judges whether an unknown ASN organization name is datacenter, hosting or VPN infrastructure. | Safety, moderation & guardrailsPrompt injection & security | |
| Coraza WAF ASN classifier — repogithub.com | — | Threat-scoring refinement in a single-binary Go web application firewall built on Coraza and OWASP CRS, where Jev judges whether an unknown ASN organization name is datacenter, hosting or VPN infrastructure. | Safety, moderation & guardrailsPrompt injection & security | |
| Dub malicious link check — appdub.co | — | Screens every new short link on the Dub link platform with Jev after a domain blacklist check, blocking phishing, malware, cloaking redirectors, gambling and adult destinations. | Safety, moderation & guardrailsPrompt injection & security | |
| is-malicious?github.com | 33 | CLI that sends source, config, build and CI files to Jev to flag hidden, deceptive or data-stealing behavior, reporting suspicious files and line ranges with a probability and reason before you run unfamiliar code. | Safety, moderation & guardrailsPrompt injection & security | |
| is-malicious? — discussionnews.ycombinator.com | — | CLI that sends source, config, build and CI files to Jev to flag hidden, deceptive or data-stealing behavior, reporting suspicious files and line ranges with a probability and reason before you run unfamiliar code. | Safety, moderation & guardrailsPrompt injection & security | |
| Jev IDSgithub.com | 2 | Network intrusion detection prototype that shows Jev one NSL-KDD flow plus five labeled examples and asks whether it is an attack and which kind, reported 4.8x faster and 3.8x cheaper than an LLM baseline. | Safety, moderation & guardrailsPrompt injection & security | |
| Jev IDS — appjev-ids.github.io | — | Network intrusion detection prototype that shows Jev one NSL-KDD flow plus five labeled examples and asks whether it is an attack and which kind, reported 4.8x faster and 3.8x cheaper than an LLM baseline. | Safety, moderation & guardrailsPrompt injection & security | |
| Jev Prompt Sentrygithub.com | — | Reverse proxy in front of Anthropic's Messages API that screens each request with one batched Jev call on four safety questions; over 1,650 guard calls it had 0 false positives on 1,590 rows and 9.1% missed attacks. | Safety, moderation & guardrailsPrompt injection & security | |
| Jev Security Scangithub.com | 10 | Standard-library Python scanner that reviews Agent Skills, MCP configs, and MCP source for prompt injection, tool poisoning, credential access, and exfiltration before install, with file and line evidence. | Safety, moderation & guardrailsPrompt injection & security | |
| Jev-AVgithub.com | 2 | Desktop antivirus and threat-triage scanner that extracts features such as entropy, PE imports, macros, and process lineage from files, processes, and URLs, then applies quarantine or review rules. | Safety, moderation & guardrailsPrompt injection & security | |
| jev-cvssgithub.com | 2 | Python scripts that score vulnerability descriptions with CVSS v3.0, v3.1 or v4.0: Jev selects each metric value and the numeric scores are computed in code per the FIRST specifications. | Safety, moderation & guardrailsPrompt injection & security | |
| jev-edgegithub.com | 2 | Admission control for LLM-backed services at the gateway, with a three-layer prompt-injection and abuse filter for nginx, OpenResty, Envoy, Cloudflare Workers, and other proxies; fail-open and cached. | Safety, moderation & guardrailsPrompt injection & security | |
| jev-gitgithub.com | 2 | Rust Git pre-commit and pre-push hook, run as git jev, that has Jev screen staged diffs for unredacted secrets, prompt-injection text and destructive commands in about 80 ms p50. | Safety, moderation & guardrailsPrompt injection & security | |
| Jev-JITx.com | — | Demo of an uncensored agent told to survive by disabling its deletion file or blackmailing an admin; Jev-JIT blocks each attempt until the agent gives up. | Safety, moderation & guardrailsPrompt injection & security | |
| Jevegis — appjevegis.vercel.app | — | Guardrails API for LLM apps that checks inputs, outputs and retrieved documents for prompt injection, jailbreaks, credential or PII leaks, malicious code and unsafe content in one Jev request. | Safety, moderation & guardrailsPrompt injection & security | |
| Jevegis — sdkgithub.com | — | Guardrails API for LLM apps that checks inputs, outputs and retrieved documents for prompt injection, jailbreaks, credential or PII leaks, malicious code and unsafe content in one Jev request. | Safety, moderation & guardrailsPrompt injection & security | |
| jevknowsgithub.com | — | Prompt-injection guard for coding agents that screens web fetches and file reads with Jev and blocks the load when agent-directed malicious instructions are found. | Safety, moderation & guardrailsPrompt injection & security | |
| laya-browser-guardgithub.com | — | Security & Guardrails: A passive, privacy-first Chrome security copilot that combines deterministic browser-visible checks with local Laya and official Jev typed decisions. It evaluates redacted evidence from scripts, resources, forms, headers, and DOM signals, then explains investigation priority without attacking the target. | Safety, moderation & guardrailsPrompt injection & security | |
| Pentest-Swarm-AI Jev scoringgithub.com | 2,521 | Autonomous pentest swarm that can use Jev as a false-positive filter and to score candidate attack paths in real time, pursuing the best-graded strategy first. | Safety, moderation & guardrailsPrompt injection & security | |
| Prompt Rejector — docsgithub.com | 2 | MCP and HTTPS scanner that screens prompts, skill files and MCP tool descriptions for injected instructions before an agent acts on them, combining deterministic checks, Jev judgments and a reasoning model. | Safety, moderation & guardrailsPrompt injection & security | |
| Responsible AI Harnessgithub.com | 1 | Assessment harness that tests AI models and agents for prompt injection, secret or PII leakage, unsafe tool use and policy bypass, combining hard rules with an optional Jev judge and a checksummed evidence report. | Safety, moderation & guardrailsPrompt injection & security | |
| Safer with Jev — appsafer-with-jev.com | — | HTTP inspection gateway on a Neon Function that has Jev judge a request body for prompt injection or unsafe content, then passes, reviews or blocks it and can forward the same bytes to a caller-chosen HTTPS URL. | Safety, moderation & guardrailsPrompt injection & security | |
| SecureAI-Scan AI014 rulegithub.com | — | Rule in SecureAI-Scan, an offline scanner for LLM, MCP and RAG vulnerabilities, that flags Python code running a dangerous action such as subprocess.run on a TypeSafe confidence score alone without an allowlist. | Safety, moderation & guardrailsPrompt injection & security | |
| SecureAI-Scan AI014 rule — npmwww.npmjs.com | — | Rule in SecureAI-Scan, an offline scanner for LLM, MCP and RAG vulnerabilities, that flags Python code running a dangerous action such as subprocess.run on a TypeSafe confidence score alone without an allowlist. | Safety, moderation & guardrailsPrompt injection & security | |
| SecureAI-Scan AI014 rule — repogithub.com | — | Rule in SecureAI-Scan, an offline scanner for LLM, MCP and RAG vulnerabilities, that flags Python code running a dangerous action such as subprocess.run on a TypeSafe confidence score alone without an allowlist. | Safety, moderation & guardrailsPrompt injection & security | |
| traffic-guardgithub.com | 1 | Zero-dependency reverse-proxy traffic classifier and bot mitigator for Node.js and Python that runs local header-order, velocity, and honeypot checks in under 100 microseconds. | Safety, moderation & guardrailsPrompt injection & security |
Content moderation64 of 64 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| 1940s.nyc story moderation experimentgithub.com | — | Local experiment for the 1940s.nyc street-view site that predicts whether a human moderator would approve a user story, asking Jev one yes/no question per moderation rule and measuring agreement with real past decisions. | Safety, moderation & guardrailsContent moderation | |
| 1940s.nyc story moderation experiment — app1940s.nyc | — | Local experiment for the 1940s.nyc street-view site that predicts whether a human moderator would approve a user story, asking Jev one yes/no question per moderation rule and measuring agreement with real past decisions. | Safety, moderation & guardrailsContent moderation | |
| 1940s.nyc story moderation experiment — repogithub.com | — | Local experiment for the 1940s.nyc street-view site that predicts whether a human moderator would approve a user story, asking Jev one yes/no question per moderation rule and measuring agreement with real past decisions. | Safety, moderation & guardrailsContent moderation | |
| Alita AI spam filtergithub.com | — | Per-chat AI spam filter for a Go Telegram group-management bot, judging each message with Jev off the update path with code-owned questions and one bounded retry. | Safety, moderation & guardrailsContent moderation | |
| Alita AI spam filter — docsalita-docs.divkix.me | — | Per-chat AI spam filter for a Go Telegram group-management bot, judging each message with Jev off the update path with code-owned questions and one bounded retry. | Safety, moderation & guardrailsContent moderation | |
| Battlesnake Arena Jev moderationgithub.com | — | Content moderation on the Battlesnake Arena competitive programming platform: Jev screens snake names, tournament text and saved-game titles, and blanks violating snake shouts from game replays. | Safety, moderation & guardrailsContent moderation | |
| Battlesnake Arena Jev moderation — repogithub.com | — | Content moderation on the Battlesnake Arena competitive programming platform: Jev screens snake names, tournament text and saved-game titles, and blanks violating snake shouts from game replays. | Safety, moderation & guardrailsContent moderation | |
| Bible Strong guestbook moderationgithub.com | — | Bible study app whose public, all-ages event guestbook is moderated by Jev, screening names and messages in any language for abuse while allowing criticism and testimony. | Safety, moderation & guardrailsContent moderation | |
| Bible Strong guestbook moderation — appbible-strong.app | — | Bible study app whose public, all-ages event guestbook is moderated by Jev, screening names and messages in any language for abuse while allowing criticism and testimony. | Safety, moderation & guardrailsContent moderation | |
| endpointgithub.com | 7 | CodeAlive-AI/mastra-jev-moderation — Input moderation for Mastra agents on TypeSafe Jev — one file · endpoint · TypeScript | Safety, moderation & guardrailsContent moderation | |
| endpointgithub.com | 1 | Infrawrench/Jeeves — Twitch/Discord moderation rules in plain English, powered by Jev and Gemini · endpoint · Rust | Safety, moderation & guardrailsContent moderation | |
| endpointgithub.com | 7 | ItisShikhar/gg-friggin-ez — Fast, drop-in profanity and toxicity screener for Node.js, powered by TypeSafe AI Jev. Catches leetspeak, character spacing, and romanized profanity across languages including Kannada, Telugu, Tamil, Hindi, and Bengali. ~50-500ms latency. · endpoint · TypeScript | Safety, moderation & guardrailsContent moderation | |
| endpointgithub.com | 14 | backmeupplz/jev_antispam_bot — Minimal grammY Telegram anti-spam bot powered by TypeSafe Jev · endpoint · TypeScript | Safety, moderation & guardrailsContent moderation | |
| endpointgithub.com | 47 | brainstormity/Jev-Moderation-Bot — A Discord moderation bot built with Python and TypeSafe AI (Jev System One). It filters spam and scam links in real time, escalates offenses automatically, and lets moderators prof · endpoint · Python | Safety, moderation & guardrailsContent moderation | |
| Eustress Engine publish moderationgithub.com | — | Layered moderation for simulations published from the Eustress Engine platform, where Jev answers a battery of typed questions over an engine-built dossier (rating, content kind, quality band) before a Grok judge and humans. | Safety, moderation & guardrailsContent moderation | |
| Eustress Engine publish moderation — repogithub.com | — | Layered moderation for simulations published from the Eustress Engine platform, where Jev answers a battery of typed questions over an engine-built dossier (rating, content kind, quality band) before a Grok judge and humans. | Safety, moderation & guardrailsContent moderation | |
| Feed slop detector extensionx.com | — | Browser extension that classifies every post on screen in real time as scam, slop, clean and similar labels, and puts a small confidence badge on each one. | Safety, moderation & guardrailsContent moderation | |
| gg-friggin-ezgithub.com | 7 | Fast, drop-in profanity and toxicity screener for Node.js, powered by TypeSafe AI Jev. Catches leetspeak, character spacing, and romanized profanity across languages including Kannada, Telugu, Tamil, Hindi, and Bengali. ~50-500ms latency. | Safety, moderation & guardrailsContent moderation | |
| gg-friggin-ez — appitisshikhar.github.io | — | Profanity and toxicity screener for Node.js that catches leetspeak, spaced-out letters and romanized profanity in languages such as Hindi, Tamil, Telugu, Kannada and Bengali in about 50-500ms. | Safety, moderation & guardrailsContent moderation | |
| Haitatsu TypeSafe spam filtergithub.com | — | Spam and phishing classification in Haitatsu, a single-binary Go email server, that asks TypeSafe for spam and phishing probabilities per message and files to Junk above per-inbox thresholds (default 0.98), with a shadow mode. | Safety, moderation & guardrailsContent moderation | |
| Haitatsu TypeSafe spam filter — repogithub.com | — | Spam and phishing classification in Haitatsu, a single-binary Go email server, that asks TypeSafe for spam and phishing probabilities per message and files to Junk above per-inbox thresholds (default 0.98), with a shadow mode. | Safety, moderation & guardrailsContent moderation | |
| Huddle message moderationgithub.com | — | Moderation check in Huddle, a Slack-like Next.js 16 team chat demo, that blocks profanity, spam, scams, harassment and hate speech before publishing by asking Jev through the Vercel AI Gateway. | Safety, moderation & guardrailsContent moderation | |
| Huddle message moderation — appnext16-team-chat.vercel.app | — | Moderation check in Huddle, a Slack-like Next.js 16 team chat demo, that blocks profanity, spam, scams, harassment and hate speech before publishing by asking Jev through the Vercel AI Gateway. | Safety, moderation & guardrailsContent moderation | |
| Huddle message moderation — repogithub.com | — | Moderation check in Huddle, a Slack-like Next.js 16 team chat demo, that blocks profanity, spam, scams, harassment and hate speech before publishing by asking Jev through the Vercel AI Gateway. | Safety, moderation & guardrailsContent moderation | |
| Infrawrench/Jeevesgithub.com | 1 | Twitch/Discord moderation rules in plain English, powered by Jev and Gemini | Safety, moderation & guardrailsContent moderation | |
| Jev Anti-Spam Botgithub.com | 14 | Self-hosted grammY Telegram bot that asks Jev whether each group message is spam, such as token shilling, DM funnels or phishing, and deletes only high-confidence matches. | Safety, moderation & guardrailsContent moderation | |
| Jev Moderation Botgithub.com | 47 | Discord bot that deletes spam and scam links in real time with escalating warnings and timeouts, and profiles members from their recent messages for scam risk, toxicity and helpfulness. | Safety, moderation & guardrailsContent moderation | |
| jev-audio-beepergithub.com | 2 | Proof of concept that finds Spanish profanity in word-timestamped audio with Jev and beeps over the matching intervals with ffmpeg without changing the audio duration. | Safety, moderation & guardrailsContent moderation | |
| LinkedIn NoSlopgithub.com | — | Chrome extension that blurs low-value LinkedIn feed posts such as engagement bait and formulaic stories, with a one-click reveal and the reason shown. | Safety, moderation & guardrailsContent moderation | |
| mail.estrogen.delivery spam filtergithub.com | — | Self-contained webmail client and server on Cloudflare Workers that classifies incoming mail with a Jev Choice over categories and treats the junk categories as spam, with an LLM fallback. | Safety, moderation & guardrailsContent moderation | |
| mail.estrogen.delivery spam filter — repogithub.com | — | Self-contained webmail client and server on Cloudflare Workers that classifies incoming mail with a Jev Choice over categories and treats the junk categories as spam, with an LLM fallback. | Safety, moderation & guardrailsContent moderation | |
| mastra-jev-moderationgithub.com | 7 | Single-file input moderation processor for Mastra agents that blocks hostile messages with one Jev request per turn, failing open behind a deadline and circuit breaker. | Safety, moderation & guardrailsContent moderation | |
| MessageWatchgithub.com | — | Red Discord Bot cog that batches recent messages in enabled channels and asks Jev three judgments about scams and hostile exchanges, posting a report to moderators when one crosses its threshold; it never deletes anything. | Safety, moderation & guardrailsContent moderation | |
| MessageWatch — repogithub.com | — | Red Discord Bot cog that batches recent messages in enabled channels and asks Jev three judgments about scams and hostile exchanges, posting a report to moderators when one crosses its threshold; it never deletes anything. | Safety, moderation & guardrailsContent moderation | |
| MillionSendgithub.com | 168 | Open-source, Resend-compatible email platform on AWS SES that samples outbound messages after sending and scores them for abuse with Jev in the background. | Safety, moderation & guardrailsContent moderation | |
| MillionSend — appmillionsend.com | — | Open-source, Resend-compatible email platform on AWS SES that samples outbound messages after sending and scores them for abuse with Jev in the background. | Safety, moderation & guardrailsContent moderation | |
| model_idgithub.com | 1 | soderlind/jev-comment-triage — Async Jev-powered WordPress comment triage: background spam, scam/phishing, and toxicity moderation that keeps comment submission fast. · model_id · PHP | Safety, moderation & guardrailsContent moderation | |
| Moderation Guardgithub.com | 1 | Comment-moderation demo where one System One call screens seven Noul hazards plus a severity Score in parallel, and code computes the verdict from strict or permissive policy thresholds. | Safety, moderation & guardrailsContent moderation | |
| Next 16 calendar Jev moderationgithub.com | — | Next.js 16 calendar and booking demo that moderates user-entered calendar text with Jev through AI SDK evaluate on Vercel AI Gateway. | Safety, moderation & guardrailsContent moderation | |
| Next 16 calendar Jev moderation — appnext16-calendar.vercel.app | — | Next.js 16 calendar and booking demo that moderates user-entered calendar text with Jev through AI SDK evaluate on Vercel AI Gateway. | Safety, moderation & guardrailsContent moderation | |
| Next 16 social media Jev moderationgithub.com | — | Next.js 16 social network demo that moderates user posts and blocks profanity with Jev through AI SDK evaluate on Vercel AI Gateway, with a 3-second timeout. | Safety, moderation & guardrailsContent moderation | |
| Next 16 social media Jev moderation — appnext16-social-media.vercel.app | — | Next.js 16 social network demo that moderates user posts and blocks profanity with Jev through AI SDK evaluate on Vercel AI Gateway, with a 3-second timeout. | Safety, moderation & guardrailsContent moderation | |
| Portfolio suggestion moderationgithub.com | — | Moderation gate for component suggestions on a developer portfolio site: heuristics catch link dumps and spam words, then two Jev Nouls judge spam and actionability before posting to Discord. | Safety, moderation & guardrailsContent moderation | |
| Portfolio suggestion moderation — appakoder.xyz | — | Moderation gate for component suggestions on a developer portfolio site: heuristics catch link dumps and spam words, then two Jev Nouls judge spam and actionability before posting to Discord. | Safety, moderation & guardrailsContent moderation | |
| Portfolio suggestion moderation — repogithub.com | — | Moderation gate for component suggestions on a developer portfolio site: heuristics catch link dumps and spam words, then two Jev Nouls judge spam and actionability before posting to Discord. | Safety, moderation & guardrailsContent moderation | |
| profanity-checkergithub.com | 1 | Cloudflare Worker API that checks text and usernames for profanity with Jev on Workers AI, catching disguised usernames such as phonetic gags and look-alike spellings. | Safety, moderation & guardrailsContent moderation | |
| Real-time chat moderationwww.reddit.com | — | Simulated live-stream chat where Jev sorts each incoming message into viewer-chosen feeds such as Questions, Feedback, or Funny, collapsing repeats and dropping spam. | Safety, moderation & guardrailsContent moderation | |
| Reply-guy stamperx.com | — | Chrome extension in which Jev reads every reply on X and hides the low-effort ones. | Safety, moderation & guardrailsContent moderation | |
| ReviewDB TypeSafe moderationgithub.com | — | Moderation module in the ReviewDB API server behind Aliucord and Vencord user-review plugins that scores reported reviews against a moderation taxonomy with Jev in a single request. | Safety, moderation & guardrailsContent moderation | |
| ReviewDB TypeSafe moderation — repogithub.com | — | Moderation module in the ReviewDB API server behind Aliucord and Vencord user-review plugins that scores reported reviews against a moderation taxonomy with Jev in a single request. | Safety, moderation & guardrailsContent moderation | |
| soderlind/jev-comment-triagegithub.com | 1 | Async Jev-powered WordPress comment triage: background spam, scam/phishing, and toxicity moderation that keeps comment submission fast. | Safety, moderation & guardrailsContent moderation | |
| Spam Detector Telegram Botgithub.com | 6 | Telegram group bot that flags spam for admin review, using Jev's spam choice, hazard probabilities and severity with no training, a local TF-IDF SVM for Persian slang, or both in hybrid mode. | Safety, moderation & guardrailsContent moderation | |
| StopSpamgithub.com | 0 | Telegram bot that removes unsolicited promotion and scams such as fake earnings, crypto giveaways, and phishing from group chats, acting only on calibrated-confidence Jev classifications. | Safety, moderation & guardrailsContent moderation | |
| Sub2API content auditgithub.com | 42,112 | Content-moderation engine in the Sub2API relay that screens traffic with Jev across 13 categories such as harassment, hate, self-harm and violence, managed from an admin risk-control page. | Safety, moderation & guardrailsContent moderation | |
| sub2api Jev content moderationgithub.com | — | Content-moderation engine in an AI API gateway fork that asks Jev 13 independent yes/no questions per text, covering harassment, hate, illicit help, self-harm, sexual content and violence. | Safety, moderation & guardrailsContent moderation | |
| sub2api Jev content moderation — repogithub.com | — | Content-moderation engine in an AI API gateway fork that asks Jev 13 independent yes/no questions per text, covering harassment, hate, illicit help, self-harm, sexual content and violence. | Safety, moderation & guardrailsContent moderation | |
| Telegram Guard Botgithub.com | 5 | Chinese Telegram group-management bot with join verification and layered anti-spam, whose AI context check can use Jev as the primary text classifier with an LLM as backup; on 31 local samples it matched DeepSeek at 0.8. | Safety, moderation & guardrailsContent moderation | |
| tg-spam Jev checkergithub.com | 446 | TG-Spam, a self-hosted Telegram anti-spam bot and library, adds a Jev spam checker that judges each message with a typed question alongside its other detectors. | Safety, moderation & guardrailsContent moderation | |
| tg-spam Jev checker — apptg-spam.umputun.dev | — | TG-Spam, a self-hosted Telegram anti-spam bot and library, adds a Jev spam checker that judges each message with a typed question alongside its other detectors. | Safety, moderation & guardrailsContent moderation | |
| toXic Filtergithub.com | 5 | Chrome extension that classifies posts on X with Jev and hides the ones matching categories you choose, such as discrimination, taunting, hate or cynicism, behind a revealable overlay. | Safety, moderation & guardrailsContent moderation | |
| TrustGate — apptrustgate-mu.vercel.app | — | Trust and safety gate for indie media that decides publish, soft-hold or block for a post, with a harm severity score and checks for hate or harassment, sensitive misinformation and spam or scams. | Safety, moderation & guardrailsContent moderation | |
| Tweet 911github.com | 3 | Chrome extension and Cloudflare Workers API that score X posts and replies in real time for AI writing, porn solicitation and paraphrase bots, using the author profile, post text and parent tweet. | Safety, moderation & guardrailsContent moderation | |
| Twitter Jev Guardgithub.com | 10 | Chrome and Edge extension that uses Jev to flag low-quality, spam and promotional posts on the X timeline, overlaying a translucent STOP or AD watermark with the probability on the post text. | Safety, moderation & guardrailsContent moderation | |
| ZeroNSFWBotgithub.com | 13 | Async Rust Telegram moderation bot that bans NSFW advertisers; images are judged locally, while Jev powers optional bio, topic, ad-guard and language checks on text. | Safety, moderation & guardrailsContent moderation |
Tool & action gates38 of 38 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| ActionGategithub.com | 2 | Runtime authorization gateway for agent tool calls that combines deterministic policy with Jev, binds approval to the exact action, and refuses expired, changed, or replayed permits over MCP and HTTP. | Safety, moderation & guardrailsTool & action gates | |
| agent-chaperone — siteagentchaperone.dev | — | Firewall that screens an AI agent's tool calls before they run and tool results before the agent reads them, via an MCP proxy and a hooks adapter, with thresholds in a policy file and a non-blocking shadow mode. | Safety, moderation & guardrailsTool & action gates | |
| AgentBeam Jev action judginggithub.com | 11 | Optional check in the AgentBeam local security CLI for coding agents that sends redacted proposed tool actions to Jev and allows, reviews or denies them in observe or enforce mode. | Safety, moderation & guardrailsTool & action gates | |
| ai_sdkgithub.com | 3 | caiovicentino/jev-shield — Semantic MCP firewall powered by Jev — screens every tool call, tool result, and tool description with calibrated System One verification. 94% block recall, 0 false positives, ~$0.00002/check. · ai_sdk · JavaScript | Safety, moderation & guardrailsTool & action gates | |
| CeamKrier/semantic-firewallgithub.com | 1 | Semantic firewall for LLM agents: tool calls gated by TypeSafe Jev (System One decision model via OpenRouter) + deterministic policy. PoC with corpus, stability eval, baseline, results. | Safety, moderation & guardrailsTool & action gates | |
| construct-auto-classifier — writeupfamelos.com | — | Safety gate in front of coding agents' shell tools (OpenCode, Antigravity) that applies structural rules, then has Jev or a chat model judge a command's reversibility and blast radius before it runs. | Safety, moderation & guardrailsTool & action gates | |
| endpointgithub.com | 2 | AkashPriyadarshii/jev-git — Sub-second Git pre-commit & pre-push semantic reflex gate powered by TypeSafe AI Jev · endpoint · Rust | Safety, moderation & guardrailsTool & action gates | |
| endpointgithub.com | 3 | ClemensSchartmueller/jev-guard — High-speed, cross-agent safety gate plugin for Claude Code, Codex CLI, and Antigravity. · endpoint · Go | Safety, moderation & guardrailsTool & action gates | |
| endpointgithub.com | 12 | Nyarlathoteppppp/pi-heed — Runtime constraints for the pi coding agent: checks every side-effecting tool call against what you said, before it runs. Powered by TypeSafe Jev. · endpoint · TypeScript | Safety, moderation & guardrailsTool & action gates | |
| endpointgithub.com | 9 | TannerMidd/specpi-jev-guard — Checks risky shell and file commands before your agent runs them. · endpoint · JavaScript | Safety, moderation & guardrailsTool & action gates | |
| endpointgithub.com | 19 | anpicasso/hermes-jev-approvals — PoC: TypeSafe Jev as the reviewer for Hermes Agent smart command approvals. 8.7x faster, 4.4x fewer prompts, measured on 153 real commands. Approvals only. · endpoint · Python | Safety, moderation & guardrailsTool & action gates | |
| endpointgithub.com | 3 | godspede/construct-auto-classifier — Effect-based safety gate for AI coding agents' shell commands (OpenCode, Antigravity): fast structural rules, then TypeSafe's Jev or a chat model judges what a command does. Certified with Jev at zero dangerous commands allowed. · endpoint · TypeScript | Safety, moderation & guardrailsTool & action gates | |
| fx auto-mode safety reviewer — benchmarkx.com | — | Safety reviewer that checks every command in the fx coding agent's auto mode; Jev benchmarked up to 18x faster at p95 and more accurate than the GPT Luna model it replaces. | Safety, moderation & guardrailsTool & action gates | |
| fx Jev permission reviewergithub.com | — | Optional permission reviewer in the fx Zig coding agent: with review_model set to Jev, the policy, context and pending action go to TypeSafe or Vercel AI Gateway and the Choice becomes the permission decision. | Safety, moderation & guardrailsTool & action gates | |
| fx Jev permission reviewer — repogithub.com | — | Optional permission reviewer in the fx Zig coding agent: with review_model set to Jev, the policy, context and pending action go to TypeSafe or Vercel AI Gateway and the Choice becomes the permission decision. | Safety, moderation & guardrailsTool & action gates | |
| fx Jev permission reviewer — websitefx.sh | — | Optional permission reviewer in the fx Zig coding agent: with review_model set to Jev, the policy, context and pending action go to TypeSafe or Vercel AI Gateway and the Choice becomes the permission decision. | Safety, moderation & guardrailsTool & action gates | |
| godspede/construct-auto-classifiergithub.com | 3 | Effect-based safety gate for AI coding agents' shell commands (OpenCode, Antigravity): fast structural rules, then TypeSafe's Jev or a chat model judges what a command does. Certified with Jev at zero dangerous commands allowed. | Safety, moderation & guardrailsTool & action gates | |
| grok-jev-guardgithub.com | — | Security & Guardrails: \grok-jev-guard\ sits immediately before a meaningful Grok Bot tool sequence. It receives a compact description of the pending operation and returns one explicit action | Safety, moderation & guardrailsTool & action gates | |
| Interlockgithub.com | — | Capability kernel that wraps every agent tool call, keeps real secrets out of the agent using canaries, and lets a policy file allow, ask or block based on Jev hazard scores. | Safety, moderation & guardrailsTool & action gates | |
| jev-firewallgithub.com | 1 | PreToolUse hook for Claude Code and Codex that checks every tool call with shell-aware deterministic rules and sends unmatched actions to Jev for an allow, ask or block decision that fails closed. | Safety, moderation & guardrailsTool & action gates | |
| jev-firewall — npmwww.npmjs.com | — | PreToolUse hook for Claude Code and Codex that checks every tool call with shell-aware deterministic rules and sends unmatched actions to Jev for an allow, ask or block decision that fails closed. | Safety, moderation & guardrailsTool & action gates | |
| jev-gategithub.com | 1 | Tool-call gate for Claude Code, Codex and Cursor that runs deterministic rules first, then one Jev request answering allow, ask or deny in about 400 ms, shipped with a 300-call prompt-injection attack kit. | Safety, moderation & guardrailsTool & action gates | |
| jev-gate — docseugeniughelbur.github.io | — | Tool-call gate for Claude Code, Codex and Cursor that runs deterministic rules first, then one Jev request answering allow, ask or deny in about 400 ms, shipped with a 300-call prompt-injection attack kit. | Safety, moderation & guardrailsTool & action gates | |
| jev-guard (ClemensSchartmueller)github.com | 3 | Go safety gate for Claude Code, Codex CLI, and Antigravity that intercepts shell, write, patch, and read tool calls, runs local boundary checks, then asks Jev about blast radius, reversibility, and destructiveness. | Safety, moderation & guardrailsTool & action gates | |
| jev-shieldgithub.com | 3 | Semantic MCP firewall between any stdio MCP client and server that screens tool calls, results and descriptions; on 56 author-built cases it blocked 33 of 35 attacks with no false positives, at about $0.00003 per check. | Safety, moderation & guardrailsTool & action gates | |
| jevshieldgithub.com | 4 | Sub-100ms security gate for AI agent tool calls, powered by TypeSafe's Jev (System-1) decision model. Single-request Choice/Noul/Score evaluation, dual-factor blocking matrix, calibrated-confidence routing, fail-closed parsing, zero-config local fallback. LangChain-ready. | Safety, moderation & guardrailsTool & action gates | |
| Leashgithub.com | 24 | Open-source safety and control layer between personal AI agents and their actions that stops destructive commands, secret exposure, prompt injection and unsafe tools, using Jev-backed decisions with your own TypeSafe key. | Safety, moderation & guardrailsTool & action gates | |
| Leash — appopenleash.com | — | Open-source safety and control layer between personal AI agents and their actions that stops destructive commands, secret exposure, prompt injection and unsafe tools, using Jev-backed decisions with your own TypeSafe key. | Safety, moderation & guardrailsTool & action gates | |
| mcp-jev-guardrailsgithub.com | — | Library for MSP MCP servers that screens tool calls against role allowlists, deny policies and user intent using atomic Jev Nouls, then lets code compose an allow/review/block decision. | Safety, moderation & guardrailsTool & action gates | |
| mcp-jev-guardrails — repogithub.com | — | Library for MSP MCP servers that screens tool calls against role allowlists, deny policies and user intent using atomic Jev Nouls, then lets code compose an allow/review/block decision. | Safety, moderation & guardrailsTool & action gates | |
| model_idgithub.com | 1 | CeamKrier/semantic-firewall — Semantic firewall for LLM agents: tool calls gated by TypeSafe Jev (System One decision model via OpenRouter) + deterministic policy. PoC with corpus, stability eval, baseline, results. · model_id · TypeScript | Safety, moderation & guardrailsTool & action gates | |
| model_idgithub.com | 2 | omkarghugarkar007/actiongate-jev — Runtime authorization and guardrails for AI-agent tool calls with deterministic policy and TypeSafe Jev via OpenRouter. · model_id · TypeScript | Safety, moderation & guardrailsTool & action gates | |
| open-jev-approvalsgithub.com | 3 | Go approval gate for Claude Code and Codex hooks: Jev reviews each intercepted tool call and a versioned local policy returns allow or deny, with a deny requiring positive evidence of danger. | Safety, moderation & guardrailsTool & action gates | |
| pi-heedgithub.com | 12 | Pi coding-agent extension that turns constraints you state in conversation, in English or Chinese, into a scoped policy and checks every side-effecting tool call against it with Jev before it runs. | Safety, moderation & guardrailsTool & action gates | |
| pi-heed — discussionwww.reddit.com | — | Pi coding-agent extension that turns constraints you state in conversation, in English or Chinese, into a scoped policy and checks every side-effecting tool call against it with Jev before it runs. | Safety, moderation & guardrailsTool & action gates | |
| pi-verdictgithub.com | 8 | Permission gate for the Pi coding agent in the style of Claude Code's auto mode: deterministic rules settle clear cases, and gray-zone tool calls go to a fail-closed classifier that can be Jev. | Safety, moderation & guardrailsTool & action gates | |
| Wanaku System One evaluatorgithub.com | — | Governed action proxy for AI agents that adds a TypeSafe System One evaluator engine, judging intercepted MCP tool calls and conversation history with Noul questions. | Safety, moderation & guardrailsTool & action gates | |
| Wanaku System One evaluator — appwanaku.ai | — | Governed action proxy for AI agents that adds a TypeSafe System One evaluator engine, judging intercepted MCP tool calls and conversation history with Noul questions. | Safety, moderation & guardrailsTool & action gates |
Output & claim verification6 of 6 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| AI reply detector with Laravel AI SDKwww.youtube.com | — | Chrome extension with a Laravel AI SDK backend that checks whether replies to the author's tweets were written by AI, comparing Jev against an OpenAI model on accuracy, cost and speed. | Safety, moderation & guardrailsOutput & claim verification | |
| Aletheiagithub.com | 6 | FastAPI service and Chrome extension that detects AI-generated text on article pages with several engines including Jev, which can also classify a page as original, repost or advertisement. | Safety, moderation & guardrailsOutput & claim verification | |
| GeekLink Jev Subtitle Translatorgithub.com | — | Media localization: reviews source–translation SRT pairs with one Jev Noul decision per cue and flags suspected omissions or meaning changes for human review. | Safety, moderation & guardrailsOutput & claim verification | |
| Prisma AIRS CLI redteam judgegithub.com | — | Command in the Prisma AIRS CLI that re-judges the attack outputs of a red-team scan with Jev and computes an independent attack success rate with confidence intervals and an agreement matrix against AIRS verdicts. | Safety, moderation & guardrailsOutput & claim verification | |
| Prisma AIRS CLI redteam judge — repogithub.com | — | Command in the Prisma AIRS CLI that re-judges the attack outputs of a red-team scan with Jev and computes an independent attack success rate with confidence intervals and an agreement matrix against AIRS verdicts. | Safety, moderation & guardrailsOutput & claim verification | |
| tweet-factcheckgithub.com | — | Fact-checks a tweet by retrieving evidence from Wikipedia and having Jev judge the claim-versus-evidence relationship, returning a verdict with confidence and sources, where not enough evidence is a normal outcome. | Safety, moderation & guardrailsOutput & claim verification |
Policy & compliance1 of 1 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| Draft rule checkerx.com | — | Jev sits between GPT and the author, rejecting every draft that breaks their rules; the post also covers what happened when Jev went quiet and why a checker needs a default branch for no answer. | Safety, moderation & guardrailsPolicy & compliance |
More projects & source code79 of 79 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| Agentgateway guardrail examplegithub.com | — | Webhook that rates jailbreak, harm, and data-leak risk with three Scores and blocks requests at level 2 or above. | Safety, moderation & guardrailsMore projects & source code | |
| Agentgateway guardrail example — repogithub.com | — | Webhook that rates jailbreak, harm, and data-leak risk with three Scores and blocks requests at level 2 or above. | Safety, moderation & guardrailsMore projects & source code | |
| AgentSkillsHub Jev scanner reviewgithub.com | — | Evaluation of Jev as a second pass for a regex security scanner over about 27.7K READMEs, asking whether a flagged line issues a behaviour or only cites it. | Safety, moderation & guardrailsMore projects & source code | |
| AGI Jev Detectiongithub.com | 4 | Local monitor for sandboxed LLM agents that classifies chains of actions with Jev and a Sentinel model, escalates through L1-L5 containment and keeps Neo4j forensics behind a dashboard; built at HackSpain 2026. | Safety, moderation & guardrailsMore projects & source code | |
| ai_sdkgithub.com | 1 | NemanjaManic/ci-gatekeeper-bot-jev — GitHub Action that uses Jev (TypeSafe AI via Vercel AI Gateway) to cheaply triage pull requests before expensive LLM/human review · ai_sdk · Shell | Safety, moderation & guardrailsMore projects & source code | |
| ai_sdkgithub.com | 11 | frostney/clean-code-review — Every code file in a pull request, judged against Uncle Bob's Clean Code by TypeSafe's Jev, then reviewed by Luna. Built on eve and Next.js. · ai_sdk · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| almcc/slop-lintergithub.com | 1 | Lints AI-generated code for slop using Jev, a System One model that makes fast structured decisions instead of generating text. | Safety, moderation & guardrailsMore projects & source code | |
| andrueandersoncs/jev-semantic-lintergithub.com | 2 | A CLI that checks natural-language engineering rules against the current Git | Safety, moderation & guardrailsMore projects & source code | |
| Blackrosegithub.com | 6 | Python and JavaScript decision layer for LLM apps that runs one Jev call on model input or output for jailbreak, harm severity and other checks, and returns allow, review or block with reasons and raw scores. | Safety, moderation & guardrailsMore projects & source code | |
| bojansandhaus/jev-home-assistant-sentinelgithub.com | 1 | A safety boundary for AI-assisted Home Assistant decisions, with explicit policy checks and deterministic state verification. | Safety, moderation & guardrailsMore projects & source code | |
| building-with-typesafe-jevgithub.com | 126 | Unofficial skill that teaches coding agents to build with TypeSafe AI's Jev: typed decisions, calibrated confidence, and prior art from 150+ community projects. | Safety, moderation & guardrailsMore projects & source code | |
| claude-jev-plugingithub.com | — | Security & Guardrails: TypeSafe Jev semantic guardrails for Claude Code | Safety, moderation & guardrailsMore projects & source code | |
| Dark Pattern HUDgithub.com | — | Android overlay that reads visible on-screen text through an accessibility service and shows a badge when Jev thinks a manipulative dark pattern is likely. | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 1 | AlexBabescu/ActionJev — Structured code review for GitHub and Gitea Actions, powered by TypeSafe Jev and written in Rust. · endpoint · Python | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 153 | DevMortimer/pi-warden — Guardrails for Pi built on pi-typesafe that steer the agent instead of interrupting you: Jev judges irreversible and off-task tool calls, detects stuck loops, checks unverified done claims, flags slop · endpoint · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 17 | Eriskii/ErisLint — Rust linter powered by configurable Jev rules, with a VS Code extension. · endpoint · Rust | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 5 | Reindeer-AI/pi-jev-guard — Check Pi code edits against repository Markdown rules with TypeSafe Jev · endpoint · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 5 | ShuhanSun/jev-oas-sentinel — Catch breaking API behavior hidden in OpenAPI prose with deterministic checks and TypeSafe JEV System One semantic review. · endpoint · Python | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 1 | Victor-Casado/if-ai — Plain-English pull request checks powered by Jev. One condition, a minimum confidence, one check. · endpoint · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 1 | cephalization/jev-oxlint — Digest your skills into oxlint rules powered by jev · endpoint · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 2 | fatwang2/jev-review-action — Configurable GitHub submission review and PR classification with TypeSafe Jev. No text-generation model. · endpoint · JavaScript | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 11 | huntedman/JevLint — Configurable semantic linting powered by Jev, with file-level NOUL judgments and a magic-strings plugin. · endpoint · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 339 | lakeday-org/perch — Semantic code linting with Jev · endpoint · JavaScript | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 1 | ohernandezdev/jev-pr-review — Score every file in a pull request with Jev (TypeSafe System One) and decide in code whether it can merge itself. Shadow mode: comments, never merges. · endpoint · Python | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 1 | shubhangi013/prune-review — Cost-aware PR review using TypeSafe Jev decisions before a generative reviewer. · endpoint · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 14 | tyler-dot-earth/patdown — typesafe's jev as a "fuzzy linter". give your code an ocular patdown. · endpoint · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 12 | valentynkit/jev-commit — pre-commit hook: one Jev call judges whether your commit message matches the diff, plus debug leftovers, scope creep, and a secret belt · endpoint · Python | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 10 | win4r/jev-security-scan — 使用 TypeSafe Jev 审查 Skill 与 MCP 可疑行为 \| Review Agent Skills and MCP code with Jev, static evidence, and explicit coverage gaps · endpoint · Python | Safety, moderation & guardrailsMore projects & source code | |
| endpointgithub.com | 65 | wobsoriano/oxlint-plugin-jev — Lint rules written in plain English. Oxlint finds the code, TypeSafe Jev answers the question. · endpoint · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| EpicEric/safe-shgithub.com | 1 | Static shell script analysis with Jev. | Safety, moderation & guardrailsMore projects & source code | |
| FreeChaOS reflex Jev backendgithub.com | — | Typed-judgment kernel crate in FreeChaOS, an agent OS forked from Codex CLI, with a Jev backend beside MiniCheck and ShieldGemma for grounding checks, caller-defined policy violations and action-risk scoring. | Safety, moderation & guardrailsMore projects & source code | |
| FreeChaOS reflex Jev backend — repogithub.com | — | Typed-judgment kernel crate in FreeChaOS, an agent OS forked from Codex CLI, with a Jev backend beside MiniCheck and ShieldGemma for grounding checks, caller-defined policy violations and action-risk scoring. | Safety, moderation & guardrailsMore projects & source code | |
| guardrail-chatbot-jevgithub.com | — | Security & Guardrails: It is a library, not a service. You call it, you get a verdict, and your code decides what to do. It runs in \\Python and TypeScript\\, both reading the same policy file, so the two sides of your stack cannot drift apart. Neither package has a third-party dependency. | Safety, moderation & guardrailsMore projects & source code | |
| Instar Jev signal shadowgithub.com | — | Persistent Claude Code agent framework that shadows its outbound message gate with Jev, logging agreement with pattern detectors for file paths, commands and config keys without affecting decisions. | Safety, moderation & guardrailsMore projects & source code | |
| Instar Jev signal shadow — specgithub.com | — | Persistent Claude Code agent framework that shadows its outbound message gate with Jev, logging agreement with pattern detectors for file paths, commands and config keys without affecting decisions. | Safety, moderation & guardrailsMore projects & source code | |
| jev-aligngithub.com | 4 | Verifier that checks LLM responses and agent plans for sycophancy, deception, overreach and dark patterns in one Jev call of about 800ms, returning block decisions with auditable probabilities. | Safety, moderation & guardrailsMore projects & source code | |
| jev-baselines-evalgithub.com | 5 | Pre-registered independent eval of TypeSafe Jev against a nano-class LLM, a frontier LLM, and a supervised encoder (Banking77 + CLINC150 zero-shot) | Safety, moderation & guardrailsMore projects & source code | |
| jev-engineeringgithub.com | — | Security & Guardrails: Jev Engineering: Typed Decision Systems for Reliable Agent Workflows. Paper, diagrams, and companion examples by Av1dlive. | Safety, moderation & guardrailsMore projects & source code | |
| jev-guardgithub.com | — | Security & Guardrails: Probability-scored guardrails for Claude Code: deny rule-breaking edits and unasked-for deploys, route your docs into each prompt, and check the final answer against the turn's own evidence. | Safety, moderation & guardrailsMore projects & source code | |
| jev-pii-checkergithub.com | 2 | CLI that scans text and files for PII in three layers: Jev gates on 13 PII categories and sensitivity, then regex and word segmentation locate the spans. | Safety, moderation & guardrailsMore projects & source code | |
| jev-reasoning-navigatorgithub.com | — | Security & Guardrails: En lugar de depender de heurísticas matemáticas frágiles o distancias vectoriales locales de coseno, \JEV-Reasoning-Navigator\ utiliza \\TypeSafe AI (\typesafe-sdk\)\\ como motor único y autoritativo de decisión cognitiva | Safety, moderation & guardrailsMore projects & source code | |
| jev-risk-check-providergithub.com | — | Security & Guardrails: x402check — LIVE payer-intent risk checks for x402 agent commerce: typed decisions (TypeSafe Jev), ES256-signed attestations, mainnet USDC settlement. did:web:x402check.xyz | Safety, moderation & guardrailsMore projects & source code | |
| jev-safety-gatewaygithub.com | 13 | Security & Guardrails: This project integrates Jev to provide structured decisions for its workflow. See the repository for implementation details. | Safety, moderation & guardrailsMore projects & source code | |
| jev-sentinelgithub.com | 8 | Guard for coding agents, shipped as a Pi extension and a Claude Code and Codex CLI plugin, that checks tool calls, tool outputs, and replies for prompt injection, risky approvals, leaked secrets, and task drift. | Safety, moderation & guardrailsMore projects & source code | |
| jev-slop-guardgithub.com | 3 | Jev Slop Guard — a Chrome extension that scores and stamps AI slop on your X and LinkedIn feeds as you scroll | Safety, moderation & guardrailsMore projects & source code | |
| jev-testgithub.com | — | Security & Guardrails: Prototype: AI-assisted NZQA marking from rubric criteria alone, using TypeSafe Jev for guardrails, criterion scores and confidence-based triage. NOT ENDORSED BY NZQA - CONCEPT ONLY | Safety, moderation & guardrailsMore projects & source code | |
| JevLensgithub.com | 0 | Chrome extension that runs Jev over the text and sanitized image evidence of articles and X or Threads posts, then badges, marks, dims, or collapses them based on the typed decisions. | Safety, moderation & guardrailsMore projects & source code | |
| JevSlopgithub.com | 2 | Site that fetches a public note.com article and asks Jev in one request for an overall judgment plus eight writing-quality Score axes, turning them into a transparent 0-100 AI Slop Score; not an authorship detector. | Safety, moderation & guardrailsMore projects & source code | |
| model_idgithub.com | 1 | bojansandhaus/jev-home-assistant-sentinel — A safety boundary for AI-assisted Home Assistant decisions, with explicit policy checks and deterministic state verification. · model_id · Python | Safety, moderation & guardrailsMore projects & source code | |
| model_idgithub.com | 2 | zdenham/jev-lint — Lint JavaScript and TypeScript against plain-English project conventions with Jev. · model_id · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| momus-reviewgithub.com | — | Security & Guardrails: Code review on rust and javascript (+languages soon) applications. Following lenses correctness, security, reliability, compatibility and testGap. | Safety, moderation & guardrailsMore projects & source code | |
| Nebula Jev tool suggestionsgithub.com | — | Nebula, an AI pentesting assistant, can have Jev rank its deferred tool catalog and connected MCP sources against the operator's recent messages before a turn starts. | Safety, moderation & guardrailsMore projects & source code | |
| NemanjaManic/ci-gatekeeper-bot-jevgithub.com | 1 | GitHub Action that uses Jev (TypeSafe AI via Vercel AI Gateway) to cheaply triage pull requests before expensive LLM/human review | Safety, moderation & guardrailsMore projects & source code | |
| NeuroSploit TypeSafe adjudicationgithub.com | — | NeuroSploit, a Rust pentest harness, uses Jev as a calibrated confirmation and adjudication layer, judging each finding confirmed, needs-review or rejected and scoring severity. | Safety, moderation & guardrailsMore projects & source code | |
| ohernandezdev/jev-pr-reviewgithub.com | 1 | Score every file in a pull request with Jev (TypeSafe System One) and decide in code whether it can merge itself. Shadow mode: comments, never merges. | Safety, moderation & guardrailsMore projects & source code | |
| Pi Jev Guardgithub.com | 5 | Pi extension that checks proposed code edits against the repository's Markdown rules before they are written, reporting the exact instruction and line range for each violation or blocking in enforcement mode. | Safety, moderation & guardrailsMore projects & source code | |
| riffgithub.com | 6 | Prose linter with ruff-style rule codes for Markdown, text, HTML, Word and PowerPoint files: static rules run locally in milliseconds, and each semantic rule is one Jev call returning a calibrated probability. | Safety, moderation & guardrailsMore projects & source code | |
| sdkgithub.com | 1 | EpicEric/safe-sh — Static shell script analysis with Jev. · sdk · Nix | Safety, moderation & guardrailsMore projects & source code | |
| sdkgithub.com | 1 | NorbertBodziony/guard-jev — Moderation Guard — TypeSafe Jev demo · sdk · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| sdkgithub.com | 1 | Thestral12/pr-sieve — Semantic PR gate: .jev.yml rules as TypeSafe Jev questions. Not a review bot. · sdk · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| sdkgithub.com | 2 | TinyFrontier/wince — Routes human review attention: green/yellow/red for a diff and who should look. It doesn't review the code. Built on TypeSafe. · sdk · Python | Safety, moderation & guardrailsMore projects & source code | |
| sdkgithub.com | 1 | almcc/slop-linter — Lints AI-generated code for slop using Jev, a System One model that makes fast structured decisions instead of generating text. · sdk · Python | Safety, moderation & guardrailsMore projects & source code | |
| sdkgithub.com | 2 | andrueandersoncs/jev-semantic-linter — A CLI that checks natural-language engineering rules against the current Git · sdk · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| sdkgithub.com | 652 | devagrawal09/jev-review — A staged code-review workflow and local dashboard built with TypeSafe Jev. · sdk · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| sdkgithub.com | 2 | opaielsheikh/typesafe-migration-guard — Automated database migration safety reviewer powered by TypeSafe AI (Jev System One model) · sdk · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| sdkgithub.com | 7 | raihankhan-rk/diffjury — DiffJury — TypeSafe Jev PR risk router + code review coach · sdk · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| sdkgithub.com | 2 | santos-sanz/jev-audio-beeper — Low-latency audio censorship POC using Jev typed decisions and ffmpeg. · sdk · TypeScript | Safety, moderation & guardrailsMore projects & source code | |
| sdkgithub.com | 6 | scale-venture-partners/riff — A small, fast prose linter: ruff-style rule codes for writing, backed by TypeSafe's Jev model · sdk · Python | Safety, moderation & guardrailsMore projects & source code | |
| sdkgithub.com | 1 | uberto/jev-test — Review Kotlin files against From Objects to Functions principles using Jev · sdk · Python | Safety, moderation & guardrailsMore projects & source code | |
| SkillCheckgithub.com | 0 | Sends an agent skill's text to Jev before installation and returns a verdict with category scores, without loading the skill into the agent's context; the submitted text is not redacted for secrets. | Safety, moderation & guardrailsMore projects & source code | |
| Slop Mopgithub.com | 1 | Chrome extension plus Vercel server that has Jev judge the writing quality of LinkedIn posts before you see them, folding or outlining suspected slop with the score and reasons, which you can overrule. | Safety, moderation & guardrailsMore projects & source code | |
| slopcheckgithub.com | — | Agent skill that checks an agent's prose for slop patterns and gates publishing, with an optional model layer via the Anthropic or TypeSafe Jev API for inflation, jargon, bare numbers and metaphors. | Safety, moderation & guardrailsMore projects & source code | |
| Sniff Testgithub.com | 33 | Prose linter for Markdown and plain text that runs countable regex rules locally and sends paragraphs to Jev for judgment rules such as restating closings, stacked hedges or clichés, one probability per rule. | Safety, moderation & guardrailsMore projects & source code | |
| Stop the Slopgithub.com | 10 | Chrome and Firefox extension that flags AI-generated YouTube scripts with thumbnail badges and sentence heatmaps, scored by a Cloudflare Worker that calls Jev with Gemini or Sapling as fallback. | Safety, moderation & guardrailsMore projects & source code | |
| Stop the Slop — codegithub.com | 10 | Chrome and Firefox extension that flags AI-generated YouTube scripts with thumbnail badges and sentence heatmaps, scored by a Cloudflare Worker that calls Jev with Gemini or Sapling as fallback. | Safety, moderation & guardrailsMore projects & source code | |
| TinyFrontier/wincegithub.com | 2 | Routes human review attention: green/yellow/red for a diff and who should look. It doesn't review the code. Built on TypeSafe. | Safety, moderation & guardrailsMore projects & source code | |
| tokengategithub.com | 4 | OpenAI-compatible proxy that checks each sliding window of tokens while an LLM response streams and cuts the stream before a violating token reaches the screen; its demo leaks 0 characters versus 173 post-hoc. | Safety, moderation & guardrailsMore projects & source code | |
| tripwiregithub.com | 3 | AI SDK middleware and OpenAI-compatible proxy that runs seven Jev checks on every LLM response in one ~100 ms call, with YAML policies, confidence-gated block or flag actions, and an eval CLI. | Safety, moderation & guardrailsMore projects & source code | |
| uberto/jev-testgithub.com | 1 | Review Kotlin files against From Objects to Functions principles using Jev | Safety, moderation & guardrailsMore projects & source code |
More posts & discussions13 of 13 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| Browser agent flaggerx.com | — | Live detector that runs Jev over a site's activity logs every 3 seconds and builds an average score to flag sessions driven by browser agents, costing under $0.01 per session. | Safety, moderation & guardrailsMore posts & discussions | |
| Human-or-AI writing detectorx.com | — | Lightweight detector that uses Jev to judge whether a piece of writing came from a human or an AI, fast enough to run with no visible delay. | Safety, moderation & guardrailsMore posts & discussions | |
| Jev as an alignment monitorwww.reddit.com | — | Early-access test of Jev as a monitor that grades harmful prompts and generations; on 4 public benchmarks it beat every other option tried while being the cheapest. | Safety, moderation & guardrailsMore posts & discussions | |
| Jev Detectorx.com | — | Free AI-slop detector that highlights formulaic, generated-sounding sentences in up to about 10,000 words in roughly 2 seconds. | Safety, moderation & guardrailsMore posts & discussions | |
| LinkedIn AI slop detector extensionx.com | — | Chrome extension, vibe-coded with Jev, that automatically detects AI slop in the LinkedIn feed. | Safety, moderation & guardrailsMore posts & discussions | |
| LinkedIn slop detectorx.com | — | Real-time detector that uses Jev to flag AI-generated slop in the LinkedIn feed as you scroll. | Safety, moderation & guardrailsMore posts & discussions | |
| Nitpickyx.com | — | AI-generated photo detector that zooms into faces, fingers, lettering, numbers and poses, where common sense tends to break, and has Jev judge each detail. | Safety, moderation & guardrailsMore posts & discussions | |
| Opus slop flaggerx.com | — | Uses Jev to flag the stock AI-writing patterns that Claude Opus tends to produce in its text. | Safety, moderation & guardrailsMore posts & discussions | |
| Real-time slop detectorx.com | — | Real-time detector that flags AI-generated slop in the feed while you scroll, powered by Jev. | Safety, moderation & guardrailsMore posts & discussions | |
| Sentence-level AI text detectorx.com | — | AI text detector that scans a full article and gives a sentence-by-sentence breakdown of what looks machine-written, in near real time. | Safety, moderation & guardrailsMore posts & discussions | |
| Slop Mop — demowww.reddit.com | — | Chrome extension plus Vercel server that has Jev judge the writing quality of LinkedIn posts before you see them, folding or outlining suspected slop with the score and reasons, which you can overrule. | Safety, moderation & guardrailsMore posts & discussions | |
| slop-filter — demonews.ycombinator.com | — | Chrome extension that scores every post and comment on X, LinkedIn, Reddit and YouTube for how likely it is AI-written and folds away those over your threshold, with weights you can fit from your own labels. | Safety, moderation & guardrailsMore posts & discussions | |
| YouTube AI-slop filterx.com | — | Bring-your-own-key Chrome extension that filters AI slop out of YouTube as you scroll and caches the results. | Safety, moderation & guardrailsMore posts & discussions |
More packages & releases1 of 1 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| tripwire — npmwww.npmjs.com | — | AI SDK middleware and OpenAI-compatible proxy that runs seven Jev checks on every LLM response in one ~100 ms call, with YAML policies, confidence-gated block or flag actions, and an eval CLI. | Safety, moderation & guardrailsMore packages & releases |
More guides & websites9 of 9 matches
| Resource | Stars | Description | Category | Save |
|---|---|---|---|---|
| AgentSkillsHub Jev scanner review — appagentskillshub.top | — | Evaluation of Jev as a second pass for a regex security scanner over about 27.7K READMEs, asking whether a flagged line issues a behaviour or only cites it. | Safety, moderation & guardrailsMore guides & websites | |
| Instar Jev signal shadow — appinstar.sh | — | Persistent Claude Code agent framework that shadows its outbound message gate with Jev, logging agreement with pattern detectors for file paths, commands and config keys without affecting decisions. | Safety, moderation & guardrailsMore guides & websites | |
| Jev Detector — appjevdetector.com | — | Free AI-slop detector that highlights formulaic, generated-sounding sentences in up to about 10,000 words in roughly 2 seconds. | Safety, moderation & guardrailsMore guides & websites | |
| jev-pii-checker — docscoo-quack.github.io | — | CLI that scans text and files for PII in three layers: Jev gates on 13 PII categories and sensitivity, then regex and word segmentation locate the spans. | Safety, moderation & guardrailsMore guides & websites | |
| JevSlop — appjevslop.pages.dev | — | Site that fetches a public note.com article and asks Jev in one request for an overall judgment plus eight writing-quality Score axes, turning them into a transparent 0-100 AI Slop Score; not an authorship detector. | Safety, moderation & guardrailsMore guides & websites | |
| rh-guard — site24601.github.io | — | Coding-agent hook for Claude Code, Cursor, Codex, Pi and others that blocks reward hacking such as tampering with graders or hidden tests, combining structural denies with a Jev risk sidecar. | Safety, moderation & guardrailsMore guides & websites | |
| Scam Shield — appscam-shield-seven-ecru.vercel.app | — | Scam text-message checker where code extracts link and sender signals, Jev answers eleven narrow questions in one call, and a pure policy function turns them into an explainable weighted risk score. | Safety, moderation & guardrailsMore guides & websites | |
| siegevnmoorthy.github.io | — | SIEGE: 200 people vs one agent. A typed action gate (TypeSafe System One) that learns from every breach, evaluated by W&B Weave, hardened by a defender loop. Built at CoreWeave Hacks: Agent Loops 2026. | Safety, moderation & guardrailsMore guides & websites | |
| Stop the Slop — appmauropello.github.io | — | Chrome and Firefox extension that flags AI-generated YouTube scripts with thumbnail badges and sentence heatmaps, scored by a Cloudflare Worker that calls Jev with Gemini or Sapling as fallback. | Safety, moderation & guardrailsMore guides & websites |
No matches this time
Try a broader search or reset your filters to explore the full collection.
Showing 244 of 244 resources