questions: { correctness: boolean, security: boolean, reliability: boolean, // +2 }
jev-review
A staged code-review workflow and local dashboard built with TypeSafe Jev.

Every project here really calls Jev in its source · refreshed daily
questions: { correctness: boolean, security: boolean, reliability: boolean, // +2 }
A staged code-review workflow and local dashboard built with TypeSafe Jev.
questions: { s0: boolean, s1: boolean, }
Lint rules written in plain English. Oxlint finds the code, TypeSafe Jev answers the question.
questions: { noul: boolean, }
typesafe's jev as a "fuzzy linter". give your code an ocular patdown.
questions: { unsolicited_promotion: boolean, profile_bait: boolean, scam_or_phishing: boolean, // +5 }
Minimal grammY Telegram anti-spam bot powered by TypeSafe Jev
questions: { instructions: boolean, }
Runtime constraints for the pi coding agent: checks every side-effecting tool call against what you said, before it runs. Powered by TypeSafe Jev.
questions: { blocking: boolean, }
Input moderation for Mastra agents on TypeSafe Jev — one file
questions: { is_toxic: boolean, is_profane: boolean, severity: score, // +2 }
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.
questions: { wants_refund: boolean, intent: choice, }
Runtime authorization and guardrails for AI-agent tool calls with deterministic policy and TypeSafe Jev via OpenRouter.
questions: { file_removal: score, credential_theft: score, data_exfiltration: score, // +5 }
Static shell script analysis with Jev.
questions: { uses: boolean, }
Configurable GitHub submission review and PR classification with TypeSafe Jev. No text-generation model.
questions: { should_review: boolean, risk: choice, route: choice, // +1 }
GitHub Action that uses Jev (TypeSafe AI via Vercel AI Gateway) to cheaply triage pull requests before expensive LLM/human review
questions: { condition: choice, }
Plain-English pull request checks powered by Jev. One condition, a minimum confidence, one check.
baseUrl = 'https://api.typesafe.ai/v1',
Semantic code linting with Jev
typesafe: { host: "api.typesafe.ai" },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
base_url: str = "https://api.typesafe.ai",
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
const ENDPOINT: &str = "https://api.typesafe.ai/v1/systemone";
Rust linter powered by configurable Jev rules, with a VS Code extension.
const response = await fetch("https://api.typesafe.ai/v1/systemone", {Configurable semantic linting powered by Jev, with file-level NOUL judgments and a magic-strings plugin.
"https://api.typesafe.ai/v1/systemone", data=body,
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.
import { experimental_evaluate as evaluate } from 'ai';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.
typesafeBaseUrl: "https://api.typesafe.ai/v1",
Checks risky shell and file commands before your agent runs them.
response = await this.fetcher("https://api.typesafe.ai/v1/systemone", {Check Pi code edits against repository Markdown rules with TypeSafe Jev
from typesafe_sdk import AsyncTypeSafeClient, Choice, RetryPolicy, TypeSafeError
A small, fast prose linter: ruff-style rule codes for writing, backed by TypeSafe's Jev model
DEFAULT_BASE_URL = "https://api.typesafe.ai"
pre-commit hook: one Jev call judges whether your commit message matches the diff, plus debug leftovers, scope creep, and a secret belt
import { choice, noul, score, TypeSafeClient } from "@typesafe-ai/sdk";DiffJury — TypeSafe Jev PR risk router + code review coach
ENDPOINT = "https://api.typesafe.ai/v1/systemone"
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
} from "@typesafe-ai/sdk";
A CLI that checks natural-language engineering rules against the current Git
import { experimental_evaluate as evaluate } from 'ai';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.
DefaultBaseURL = "https://api.typesafe.ai/v1/systemone"
High-speed, cross-agent safety gate plugin for Claude Code, Codex CLI, and Antigravity.
TYPESAFE_BASE_URL Jev API root (default https://api.typesafe.ai)
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.
import { choice, TypeSafeClient } from "@typesafeai/sdk";Automated database migration safety reviewer powered by TypeSafe AI (Jev System One model)
DEFAULT_ENDPOINT = "https://api.typesafe.ai/v1/systemone"
Catch breaking API behavior hidden in OpenAPI prose with deterministic checks and TypeSafe JEV System One semantic review.
from typesafe_sdk import Choice, Noul, RetryPolicy, Score, TypeSafeClient
Routes human review attention: green/yellow/red for a diff and who should look. It doesn't review the code. Built on TypeSafe.
.parse(models.data.find((item) => item.id === "typesafe-ai/jev"));
Lint JavaScript and TypeScript against plain-English project conventions with Jev.
let resp = ureq::post("https://api.typesafe.ai/v1/systemone")Sub-second Git pre-commit & pre-push semantic reflex gate powered by TypeSafe AI Jev
#[arg(long, default_value = "https://api.typesafe.ai/v1/systemone")]
Structured code review for GitHub and Gitea Actions, powered by TypeSafe Jev and written in Rust.
from typesafe_sdk import SystemOneResponse
Lints AI-generated code for slop using Jev, a System One model that makes fast structured decisions instead of generating text.
MODEL = "typesafe/jev-1.13"
A safety boundary for AI-assisted Home Assistant decisions, with explicit policy checks and deterministic state verification.
jevModel: process.env.JEV_MODEL ?? "typesafe/jev-1.13",
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.
url: `${process.env.OXLINT_JEV_BASE_URL ?? DEFAULT_BASE_URL}/v1/systemone`,Digest your skills into oxlint rules powered by jev
assert!(headers.starts_with("post /v1/systemone "));Twitch/Discord moderation rules in plain English, powered by Jev and Gemini
import { TypeSafeClient, noul, score } from "@typesafe-ai/sdk";Moderation Guard — TypeSafe Jev demo
TYPESAFE_URL = "https://api.typesafe.ai/v1/systemone"
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.
import { noul, score, TypeSafeClient } from '@typesafe-ai/sdk';Low-latency audio censorship POC using Jev typed decisions and ffmpeg.
this.baseUrl = opts.baseUrl ?? process.env.JEV_BASE_URL ?? "https://api.typesafe.ai/v1";
Cost-aware PR review using TypeSafe Jev decisions before a generative reviewer.
$columns['jev_triage'] = __( 'Jev', 'jev-comment-triage' );
Async Jev-powered WordPress comment triage: background spam, scam/phishing, and toxicity moderation that keeps comment submission fast.
import { noul } from "@typesafe-ai/sdk";Semantic PR gate: .jev.yml rules as TypeSafe Jev questions. Not a review bot.
The `jev` decorator discards confidence, so this calls Jev through `typesafe_sdk` directly.
Review Kotlin files against From Objects to Functions principles using Jev