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

typesafe-cli (geilt)

Stdlib-only Python CLI plus agent skill for asking System One typed Choice, Score and Noul questions from the shell.

Agent skillbeta
View on GitHub ↗

Overview

TypeSafe CLI is a small Python command-line tool and agent skill for TypeSafe System One. It evaluates a state against typed questions and prints structured answers, using only the standard library and Python 3.9+. Commands include info, models, a smoke test from the docs quickstart, and ask, which takes inline questions or state and question files and can output JSON.

The bundled SKILL.md can be copied into Claude, Codex, Grok or other Agent Skills hosts so agents know when and how to call it. The README gives practical guidance: use it when code needs a snap judgment and the answer shape is known; do not use it for generation or to replace deterministic rules; batch independent questions about one state; and read Noul 0.5 as unsure, not medium.

What the decision model decides

TypeQuestionAnswers
Yes / noDoes this convey urgency?yes · no
ChoiceUser-defined Choice question—
ScoreUser-defined Score question—

How it works

  1. Export TYPESAFE_API_KEY or pass --creds.
  2. Run typesafe ask with state and questions.
  3. Read structured answers or JSON.
  4. Agents call it through the installed skill.

Strengths

  • No dependencies beyond Python's standard library.
  • Includes an agent skill and usage guidance.
  • Does not print the key in info output.

Caveats

  • Requires a TypeSafe key.
  • Small community tool.

Getting started

git clone https://github.com/geilt/typesafe-cli.git && export TYPESAFE_API_KEY=... && ./typesafe smoke

At a glance

Type
Agent skill
Status
beta
Language
Python
Runtime
CLI, agent skill
Has code
yes

Use cases and integrations

How curated lists describe it

Frequently asked questions

What does it need?

Python 3.9+ and a TypeSafe API key.

How do agents use it?

Copy the bundled SKILL.md into the agent's skills folder.

What does Noul 0.5 mean?

Unsure, not medium, per the README.

Related projects

  • askjev

    Unofficial hosted or local Jev MCP server giving agents a quick second opinion: it picks yes/no, scale or choice and returns probabilities.

    MCP serverbetaTypeScript
  • System One Connector

    Go CLI and single-binary MCP server giving Claude Code, Codex, Hermes and pi an evaluate tool backed by Jev, d1, CLM or Laya.

    MCP serverbetaGo
  • swift-jev

    Swift library and CLI for Jev with enum-typed Choice, Noul and Score questions, plus an agent skill for coding agents.

    CLIproductionSwift
  • jev-mcp (blakestone-x)

    Python MCP server wrapping Jev as classify, score, check, match and screen tools with confidence on every answer.

    MCP serverbetaPython
  • openrouter-jev-mcp

    Python decision gateway and stdio MCP server exposing Jev via OpenRouter's alpha Decisions endpoint to Claude Code, Codex and Cursor.

    MCP serverbetaPython
  • prompt2jev

    Agent skill and CLI that converts plain requirements, LLM prompts or prompt-calling code into a typed Jev request and runnable script.

    Agent skillbetaPython