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

hermes-jev

Hermes Agent tool plugin giving bounded Jev judgments and an explicit model recommendation via TypeSafe, Cloudflare or OpenRouter.

Agent skillprototypeMIT
View on GitHub ↗

Overview

This hermes-jev is a native Hermes Agent tool plugin, not a chat-model provider. Jev supplies bounded semantic judgments and an explicit model recommendation, while the main Hermes agent keeps planning, generation, tool execution and responsibility for approvals. It supports three backends, each selected explicitly: TypeSafe's official API, Cloudflare AI and OpenRouter's alpha decisions endpoint, which rejects the jev-latest alias so a versioned model is pinned.

There is deliberately no cross-provider fallback, credential reuse, automatic retries or redirected authenticated requests, and results preserve the actual model ID returned. Version 0.1.2 is a prerelease: a live Cloudflare smoke test in Chinese passed with jev-1.13.0, but direct TypeSafe access and real-workload accuracy remain unverified. It targets Hermes 0.21.3 and includes notes on the official skill in Chinese.

What the decision model decides

TypeQuestionAnswers
ChoiceWhich model should handle this task (recommendation)?—
Yes / noBounded semantic judgments requested by the agent(inferred)yes · no

How it works

  1. Install and run hermes jev setup with one backend
  2. Agent calls the plugin's tools for judgments or model recommendations
  3. Plugin queries the chosen backend only
  4. Agent uses the answer; approvals stay with Hermes

Strengths

  • Strict, explicit backend selection
  • No silent fallbacks
  • Preserves returned model IDs

Caveats

  • Prerelease; TypeSafe path and accuracy unverified
  • Targets a specific Hermes version

Getting started

hermes plugins install ourines/hermes-jev --enable, then hermes jev setup --backend typesafe (or cloudflare / openrouter).

At a glance

Type
Agent skill
Status
prototype
Language
Python
Runtime
Hermes Agent tool plugin
Has code
yes
License
MIT

Use cases and integrations

How curated lists describe it

Frequently asked questions

Is it a chat model provider?

No, it is a tool plugin for judgments and model recommendations.

Which backends?

TypeSafe, Cloudflare AI and OpenRouter, chosen explicitly.

Is it production-ready?

No, v0.1.2 is a prerelease.

Related projects

  • cursor-jev

    Cursor plugin whose hooks ask Jev whether subagent spawns, file reads, writes and shell commands are worth taking.

    PluginbetaJavaScript
  • Helm

    Orchestrator that probes which coding agents are installed and logged in, then has Jev pick the best one for a task.

    Agent skillbetaPython
  • jev-agent-hooks

    Hooks for Claude Code, Codex and pi that use Jev to suggest a skill each turn, plus a rule gate for subagent model routing.

    HookbetaJavaScript
  • jev-codex-router-skill

    Portable Codex skill that asks Jev to choose a model and reasoning tier for independent subtasks, with confidence gating.

    Agent skillbetaPython
  • jev-conductor-router

    MCP server and CLI where Jev decides if a coding task needs a Conductor cloud workspace, how hard it is, and which harness and model run it.

    ProjectbetaTypeScript
  • jev-mcp

    Standalone MCP server for T3 Code with one jev_route tool where Jev picks a model and reasoning-effort combination for a task.

    MCP serverbetaTypeScript