Do I need an API key to try these tools?

Clicking a live link runs hosted on jevtypesafeai.com with no key, while opening index.html locally requires pasting a key into the Jev API key setting.

Which key should I use in the browser?

A key from jevtypesafeai.com works in a minute with a browser-ready endpoint, while an official TypeSafe key may be blocked by CORS unless proxied.

What does the loop detector actually output?

The loop detector grades each pasted run step as progress, repeating, stuck, or escalate and flags looping steps for copying.

What does the skill router decide for each task?

It routes each pasted task to one tool such as web search, run code, SQL query, or send email, with confidence and copyable JSON routing.

How is each tool packaged?

Each is a single dependency-free index.html. Open it, click the Jev API key button, paste a key and run it on your own logs, traces or tasks with no server and no npm install; the hosted links run the same tools on jevtypesafeai.com.

What does the client-side example do?

It posts a step from an agent run to the decision endpoint as a choice question grading the step progress, repeating, stuck or escalate, then branches in code on the returned answer, for example flagging any step that is not progress to halt a runaway loop.

What latency and licence does it have?

Jev returns a typed, calibrated decision in about 100 to 300 ms with no chat and no parsing, and the tools are MIT licensed with pull requests welcome.

CommunityJev Agents & MCP

Jev Agent Tools

Two dependency-free single-file browser tools: an agent loop detector and a skill router powered by Jev decisions.

See highlights

Open github.com

Stars: 0
Language: HTML
License: MIT
Published by: Community
Author: codaaiteam
Last verified: Oct 2026

codaaiteam/jev-agent-tools

User repository on GitHub

View on GitHub

Open-source single-file tools for AI agent builders, powered by Jev (TypeSafe AI's decision model): a loop detector that catches runaway agents and a task→tool router. No build. Use free: jevtypesafeai.com

Forks: 0
Last pushed: Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.

About this project

Small, open-source, single-file tools for people building AI agents — powered by Jev, TypeSafe AI's System One decision model. You give Jev a situation and typed questions; it returns a typed, calibrated decision (a choice, score, or yes/no probability) in 100–300 ms — no chat, no parsing, no hallucinations. Each tool is a single dependency-free index.html. Open it, paste a key, use it on your own data. Zero setup: click a "use it live" link — runs on jevtypesafeai.com, no key.

Quoted from the project's README. Read the full README.

Tagscommunity github

Works forJev for Routing Jev for Tool calling

Used inJev for Software & DevTools

Highlights

Grades agent runs step by step as progress, repeating, stuck, or escalate.
Routes tasks to one tool such as web_search, run_code, sql_query, or send_email.
Each tool is a single dependency-free index.html using a typed Jev decide call.
Stores the key only in browser localStorage with no server or install step.

Watch out

Browser tools calling a third-party Jev endpoint; keys stay in localStorage and official API browser calls may need a proxy. Copy reviewed against the repository README on 2026-10-04, not from running it.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

agent-harness-with-jev-llm

Demo of the LangChain Jev harness pattern in LangGraph: a fast classifier routes events and gates a risky tool call, leaving open-ended reasoning to a chat LLM. Entry is POST /events; output lands in JSONL files.