Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

fx

Vercel Labs' tiny Zig coding agent, listed for a typesafe_permission_reviewer builtin that runs permission decisions through Jev.

ProjectprototypeApache-2.0
View on GitHub ↗

Overview

fx is a minimal coding agent CLI from Vercel Labs, built as a single native Zig binary with no runtime, running the same on a laptop, in CI or inside an agent sandbox. Output prints into terminal scrollback like a Unix tool. It is meant to be embedded: scripts call fx ask, editors connect over the Agent Client Protocol, and JavaScript apps use the same core through libfx. Built-in tools cover files, shell, web search and subagents, and models come from Vercel AI Gateway, ChatGPT or Grok subscriptions, or OpenAI-compatible endpoints. CI enforces strict performance budgets.

According to the list summary, fx ships a typesafe_permission_reviewer builtin so the agent's permission decisions run through Jev rather than an LLM call. The README does not describe this builtin. The project is marked experimental.

What the decision model decides

TypeQuestionAnswers
Yes / noShould this agent action be permitted?(inferred)yes · no

How it works

  1. Agent proposes an action needing permission
  2. typesafe_permission_reviewer asks Jev
  3. Allow or deny based on the typed verdict

Strengths

  • Permission checks without an extra LLM call
  • Tiny, fast native agent

Caveats

  • Jev reviewer not described in the README
  • Experimental status

Getting started

curl -fsSL https://fx.sh/setup.sh | bash.

At a glance

Type
Project
Status
prototype
Language
Zig
Runtime
native CLI
Has code
yes
Has tests
yes
License
Apache-2.0

Use cases and integrations

How curated lists describe it

  • Coding agent: ships a `typesafe_permission_reviewer` builtin so the agent's permission decisions run through Jev rather than an LLM call.

    yibie/awesome-jev

Frequently asked questions

Where is Jev used in fx?

In a typesafe_permission_reviewer builtin for permission decisions, per the listing.

What is fx?

A minimal, embeddable coding agent written in Zig.

Is it stable?

No, it is marked experimental.

Related projects

  • jev-zig-cli

    Auditable Zig terminal agent pairing Codex for coding with Jev typed preflight and postflight checks under deterministic policy gates.

    CLIproductionZig
  • actiongate-jev

    Gateway that checks agent tool calls with policy plus Jev evidence, then binds approval to a single-use permit for that exact action.

    MCP serverbetaTypeScript
  • Agent Chaperone

    MCP proxy and hooks that screen agent tool calls before they run and their results before the agent reads them, using Jev judgments.

    MCP serverbetaTypeScript
  • agent-guardrail-jev

    Xano backend template where Jev judges each agent tool call as safe, needs review or dangerous, and uncertain calls wait for a human.

    ProjectbetaXanoScript
  • approval-judge-bridge

    OpenAI-compatible endpoint that answers an agent's command-approval call with APPROVE, DENY or ESCALATE from a Jev Choice, failing closed.

    ProxybetaPython
  • claude-jev-plugin

    Claude Code plugin adding Jev judgments to Bash, Write and Edit calls and their outputs; a semantic guardrail, not a sandbox.

    Pluginbeta