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

bf-jev-deep-research

Portuguese agent skill teaching Claude Code, Cursor and Codex to use Jev correctly, bundled with the full raw research behind each claim.

Agent skillproductionMIT
View on GitHub ↗

This repository is mostly documentation or write-up rather than runnable code.

Overview

bf-jev-deep-research installs, with one command, an agent skill for Claude Code, Cursor and Codex that teaches an agent how to use Jev correctly, along with the complete raw study supporting every claim in the skill: eight analysed video transcripts, 50 pages of official documentation as raw markdown, three pages from the company site and seven synthesis documents in Portuguese.

It summarizes key facts from the sources, such as RLCD training, 70 to 500 ms decisions, pricing, the three primitives, four documented composition patterns and nine known pitfalls with real cases, and verifies more than 500 internal links automatically on every change. Installation can be per project or global.

What the decision model decides

TypeQuestionAnswers
OtherHow should an agent phrase and use Choice, Score and Noul questions correctly?(inferred)—

How it works

  1. Run the setup script (project or global)
  2. Skill activates in the agent
  3. Agent consults the skill and raw study when using Jev

Strengths

  • Every claim traceable to raw sources
  • Link checking on each change
  • Works across several agents

Caveats

  • Content in Portuguese
  • Mostly documentation

Getting started

curl -fsSL https://raw.githubusercontent.com/BFLabsAI/bf-jev-deep-research/main/scripts/setup-bf-jev-deep-research.sh | bash

At a glance

Type
Agent skill
Status
production
Runtime
agent skill
Has code
yes
License
MIT

Use cases and integrations

How curated lists describe it

Frequently asked questions

What does the skill teach?

How to use Jev's primitives and patterns correctly and avoid known pitfalls.

What is bundled?

Raw docs, transcripts and synthesis documents backing each claim.

Which agents?

Claude Code, Cursor and Codex.

Related projects

  • jev-architect (BenjaminPolge)

    Skill that makes Claude Code and Codex split AI features into Jev, LLM or plain-code steps before writing code.

    Projectbeta
  • jev-browser-skill

    Readable reference skill for Claude Code and Codex where Jev picks from a numbered table of page controls, with an explainer site.

    Agent skillprototypeJavaScript
  • Building with Jev★ 155

    Agent skill for writing and improving programs that call Jev: question design, state structure, thresholds and diagnosing wrong answers.

    Agent skillproduction
  • jev-agent-toolkit

    Portable Agent Skill teaching coding agents to use Jev as a bounded judgment layer, plus an optional MCP bridge.

    MCP serverbetaTypeScript
  • 发现与本库做同一件事的 agent skill：`decision-first`

    Agent skill that makes a coding agent try Jev first for bounded judgments like classify, rank or gate, and log each attempt.

    Agent skillbetaPython
  • Agent Tool

    IDE and browser extensions to find and install agent Skills and Subagents, with an opt-in Jev check when local detection can't decide.

    Browser extensionproductionTypeScript