User repository on GitHub
Agent skill for designing, building, testing, and tuning bounded JEV decision layers
- Forks
- 1
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Agents & MCP
Jev Anything
An independent agent skill that walks a coding agent through designing a bounded Jev decision layer: contract, criteria, error behavior, fallback, and evaluation plan first, then scaffolding through TypeSafe or OpenRouter.
- Stars
- 32
- Language
- Python
- License
- MIT
- Reading time
- 2 min
- Published by
- Community
- Author
- grandamenium
- Last verified
- Oct 2026
- Languages
- PythonTypeScript
- Platforms
- CLIOpenRouter
- Capabilities
- Open sourceRequires API key
Highlights
- Teaches the agent to pin the primitive first: noul for yes/no probability, choice for one option of up to 255, score for 2–10 ordered levels.
- Ships SKILL.md, a plan template, example prompts, reference docs, and dependency-free Python and TypeScript clients.
- Installs by cloning into ~/.codex/skills or ~/.claude/skills, with the repository root as the skill folder.
- Two access paths: a TypeSafe key or OpenRouter, whose live model page is typesafe/jev-1.13.
- Says plainly that Jev returns a signal: thresholds, permissions, side effects, and escalation stay in your code.
Quickstart
bash
mkdir -p ~/.codex/skills
git clone https://github.com/grandamenium/jev-anything.git \
~/.codex/skills/jev-anythingWatch out
Copy reviewed against the repo README on 2026-10-15. It is an independent project, not an official TypeSafe or OpenRouter product, and TypeSafe publishes its own official skill at typesafe-ai/skills. The README warns that APIs and model aliases change and asks you to recheck the live docs before implementing.
Questions about this entry
Short answers drawn from the linked project or page.
An independent agent skill for designing, implementing, testing, and tuning a bounded Jev decision layer. It makes the agent define the decision contract, criteria, error behavior, fallback, and evaluation plan before writing any code.
The repository root is the skill folder: clone it to ~/.codex/skills/jev-anything for Codex or ~/.claude/skills/jev-anything for Claude Code, then reload the agent and confirm that SKILL.md exists.
noul returns a yes/no probability, choice picks one option from a fixed list of up to 255, and score places the input on an ordered rubric of 2 to 10 levels. The skill confirms which one fits your problem before scaffolding the integration.
Both work. Set TYPESAFE_API_KEY for the TypeSafe direct API with the first-party HTTP API or Python SDK, or set OPENROUTER_API_KEY and adapt the dependency-free Python or TypeScript client in scaffolding/, checking the live model page for current availability and price.
No. The README says it is independently produced, not an official TypeSafe or OpenRouter product, and notes that TypeSafe publishes its own MIT-licensed official agent skill at typesafe-ai/skills.
It turns the request into one bounded decision, confirms the primitive, explains fit, limits, privacy, cost, latency, and high-stakes concerns, produces a reviewable plan for sign-off, and only then scaffolds the integration and a blind-label evaluate-and-tune loop.
Pull the installed copy: git -C ~/.codex/skills/jev-anything pull --ff-only, using the equivalent path if your agent stores skills elsewhere.
More Jev resources like this
A native Hermes Agent tool plugin - not a chat-model provider - where Jev supplies bounded semantic judgments and an explicit model recommendation while the main agent keeps planning, generation, execution and approvals. Backends: TypeSafe, Cloudflare and OpenRouter.
Jev Agents & MCP#community#catalog-import#logicrw
Community
ourines/hermes-jevPython
ourines/hermes-jevPythonAn agent skill for Claude Code, Codex and other coding agents, with a standard-library Python CLI, 5 tested question sets and findings from 19,367 Jev calls and 90.4M metered input tokens over 4 days.
Jev Agents & MCP#community#catalog-import#logicrw
Community
marcodicesare-dev/jev-skillPython
marcodicesare-dev/jev-skillPythonA demo-MVP coding-agent skill and CLI that hunts AI and software jobs on real company sites: Chrome sees and acts, Jev scores every link and job, and the host LLM never picks what to click.
Jev Agents & MCP#community#catalog-import#logicrw
Community
hqman/jevscoutPython
hqman/jevscoutPythonClaude Code plugin that asks Jev a few typed questions before every turn and sets reasoning effort, subagent model and effort, strategy, and the skill a prompt needs; you keep Opus for the conversation.
Jev Agents & MCP#community#catalog-import#logicrw
Community
akramovic1/jev-pilotTypeScript
akramovic1/jev-pilotTypeScriptpi extension that routes every prompt to a task-appropriate model tier using four TypeSafe Jev typed judgments, with budget-aware downgrades, candidate fallback chains, and the decision recorded in the transcript.
Jev Agents & MCP#community#catalog-import#logicrw
A dependency-free MCP server that exposes TypeSafe's Jev to Claude Code, Codex, Cursor, Hermes, and other agents through the ask_jev and list_jev_models tools, so your code gets typed decisions with probabilities instead of prose to parse.
Jev Agents & MCP#community#catalog-import#logicrw
Community
codeia-academy/jev-mcpPython
codeia-academy/jev-mcpPythonFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.