Agent action approval with decision models

From-scratch research reconstruction of a small System-One decision model, used to route and gate commands in a Windows palette.

Pre-action gate and browser agent where rules decide first, then Jev or a local Jev-compatible model, emitting allow, confirm or block.

Kernel coordinating several Claude Code agents on one repo: Jev rates collision risk before each write to pass, wait or escalate.

Preliminary SHADE-Arena study of Jev as a monitor and per-action gate for covert agent sabotage, compared with Gemini 2.5.

Zero-dependency Python SDK for deterministic gates over calibrated model answers, with LangChain, OpenAI and Claude agent hooks.

Natural-language CLI where Jev picks which installed 'reflex' programs a sentence calls, running them only when sure enough.

Library, CLI and MCP server where an LLM states each step's outcome and Jev decides element, action, value and done/blocked/irreversible.

Governance proxy between Jev or local Laya and agents that turns probabilities into allow, ask, deny, keep, truncate or drop actions.

PRAXEON runtime supervisor for autonomous agents: actions are checked by policy, a LAYA or TypeSafe semantic judge, signed and sandboxed.

LangChain and LangGraph middleware that screens tool outputs for prompt injection and gates risky tool calls with Jev.

Vercel AI SDK layer that uses Jev to auto-approve, ask a human or block each tool call, and to prune irrelevant tools before a turn.

Unofficial TypeScript monorepo of Jev tools: a tool-call guard, a GitHub issue triage Action, a bulk JSONL/CSV sorter and a shared client.

Claude Code, Codex and pi plugin that hands non-writing agent steps to Jev and escalates writing or unsure steps to the LLM.

Zero-dependency Node library with MCP and ACP adapters adding Jev decisions and gated computer use beside any LLM agent.

Pre-install npm security gate where TypeSafe System One judges lifecycle scripts for intent, threat, secret access and remote execution.

Unix CLI for typed semantic decisions in pipelines and CI: Jev predicates, routes, scores and filters with predictable exit codes.

Decision-to-action harness that puts Jev between an agent and your data to find the file, check the claim and permit actions.

Pi-based coding agent where a small judge (Jev by default) answers yes/no, choice and score questions at 38 decision points per turn.

Python policy-as-code SDK that runs Noul, Choice and Score rules defined in YAML through Jev, with audit mode and fail-closed options.

Research proposal-review contract: an LLM proposes one action, Jev answers four narrow questions, code produces evidence, applies nothing.

Pi agent harness where Jev routes the model tier, picks runbook sections, gates tool calls and verifies reports on Neon Postgres.

Open protocol for decision-based agents: apps expose evidence and typed decisions, Jev assesses, and app code guards commits.

Zero-dependency agent harness that sends each loop fork (act?, which tool, risk, done?) to a decision model and keeps the LLM for writing.

Small web demo that takes a planned agent tool call and context and returns allow, ask_human or deny with Jev risk scores.

MoonBit and TypeScript Jev playground: gomoku, shell-command risk gate, ESLint plugin, a tiny Jev-condition language and measured reports.

AWS Strands Labs' 2B decision model: Qwen3.5-2B with a pointer head scoring options and calibrated confidence, plus a CLI.

Local Jev alternative on zero-shot NLI models with temperature scaling and conformal abstention; reports ECE 0.170 to 0.071, offline.

Agent skill that sends closed coding-agent judgments to Jev before asking the user, before risky commands and after failures.

Open 0.8B and 2B Qwen3.5 decision models with a local systemone-compatible server, MCP tools, agent skills and a Claude Code guard.

Small one-pass classifier that labels shell commands as read-only, reversible or destructive before an agent runs them.