Decisions from the command line (X)

Post on X

Views: 117,027
Likes: 1,239
Reposts: 58

Read from X on 24 Sep 2026. Counts change daily.

Chris Tate’s post announces that every agent harness can now use Jev through ai-cli, installed with npm install -g ai-cli. An agent harness is the tool or framework that runs an agent’s steps.

The post lists three question types: yes/no questions, choosing between options, and scoring against your own criteria. The package is Vercel Labs’ ai command, whose evaluate subcommand takes the input on stdin and uses Jev as its default evaluation model.

Open on X

View the post on X

PrimitivesChoice Score Noul

Use casesTyped tool dispatch

Jev Ultrafast (Browser Use)

Browser Use's fast browser agent where Jev picks the operation and the DOM element in a single request; a small LLM is only invoked to write text when typing is required. Demoed on a Google Flights search.

browser-useOpen

19k starsFlight search 7.1 s

ProjectTyped tool dispatch+2

typesafe-ai/skills (official TypeSafe agent skill)

Official agent skill for Claude Code, Codex and other agent environments; installs via claude plugin marketplace add typesafe-ai/skills or npx skills add typesafe-ai/skills. MIT licensed.

TypeSafe AI2k starsOpen

ToolAgent routing and skill selection+1

jkudish/jev-mcp

MCP server that gives Claude Code, Codex, Amp, OpenCode and other agents ten Jev judgment tools, each returning typed probabilities they can branch on. The author calls it early software.

jkudishOpen

320 starsPer call 150-500 ms

ToolAgent routing and skill selection+2

Agent Desktop

Rust desktop automation that reads apps through the OS accessibility tree instead of pixels. Its jev-desktop skill takes a plain-language goal and has Jev pick one operation and one target element per turn, so the 150-element tree never enters the agent's context.

lahfirOpen

1.6k starsToken reduction 78-96%

ProjectTyped tool dispatch+1

typesafe-computer-use

macOS computer-use loop that OCRs the screen, classifies the next action with Jev, then clicks. Author reports roughly $0.0002 per step.

Aaron Levin (awlevin)Open

935 starsCost/decision $0.0002, 155xLatency 0.13-0.38s

ProjectReal-time control+1

milind-soni/tiptour-macos

A macOS menu bar companion with two modes: type a task and Jev picks what to click from a local OmniParser running on CoreML, looping for up to 12 actions, or talk to Gemini Live, which takes one desktop action per turn.

Milind S (@milindlabs)Open

662 starsAction limit 12 actions

ProjectTyped tool dispatch

Menu