Agent Plugins Marketplace
← All plugins

jev

v1.0.3

System One sweeps over the jev-latest model: ground text against sources, pick browser elements, tag mail and rows.

CodexClaude Code1 Skill

By Nathan Wong1 GitHub starsUpdated 1 hour ago

Directory evidence

Runtimes
Codex and Claude Code
Parsed components
1 skill or MCP entry
Source updated
Sep 27, 2026
Manifest status
Canonical path parsed

The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →

Install jev for Codex and Claude Code

Installs for the current user
codex plugin marketplace add thelogicmatrix/lm-tools
codex plugin marketplace upgrade lm-tools
codex plugin add jev@lm-tools

Paste and run these commands in a terminal with Codex. They add and refresh the lm-tools catalog, then install this plugin.

Compatibility: the page URL and API slug “jev-3” remain stable.

  • Codex: jev-3@agent-plugin-marketplace → jev@lm-tools

The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.

Get the source manually
git clone https://github.com/thelogicmatrix/lm-tools

Clone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is jev/.

Plugin files

jev/
├── .codex-plugin/plugin.json
├── .claude-plugin/plugin.json
└── skills/jevchecker/SKILL.md

Included Skills1

jevcheckerskills/jevchecker/SKILL.md

Sweeps a large body of text for a nuanced criterion and returns only the parts worth reading, so a Claude context sees the shortlist instead of the whole body. Use when the corpus is too big or too costly to read in full and the criterion is one a regex cannot express, such as grounding hundreds of generated bullets against the sources they came from. The shipped resume sweep is a calibration case, since one resume is small enough to read. Never use on customer data, because the whole body is posted to OpenRouter. Do not use when the body is small enough to just read, when the question compares one part against another, or when a verdict is wanted rather than a shortlist.

Plugin manifests2

jev/.codex-plugin/plugin.json
{
  "name": "jev",
  "version": "1.0.3",
  "description": "System One sweeps over the jev-latest model: ground text against sources, pick browser elements, tag mail and rows.",
  "author": {
    "name": "Nathan Wong"
  },
  "skills": "./skills/",
  "interface": {
    "displayName": "jev",
    "shortDescription": "Cheap sweeps over large bodies of text",
    "longDescription": "jevchecker returns the parts of a large body worth reading against a nuanced criterion. The scripts folder holds the other sweeps (jevclick, jevmail, jevclassify, jevdrift) and the shared lib.",
    "developerName": "Nathan Wong",
    "category": "Developer Tools",
    "capabilities": [
      "Read"
    ],
    "defaultPrompt": "Sweep this body for the parts worth reading."
  }
}
jev/.claude-plugin/plugin.json
{
  "name": "jev",
  "version": "1.0.3",
  "description": "jev - Nathan's System One sweeps over the jev-latest model: jevchecker grounds generated text against its sources, jevclick picks a browser element from a snapshot, jevmail tags a mailbox, jevclassify labels repos. One shared lib, one key in ~/.jev.env.",
  "author": {
    "name": "Nathan Wong"
  }
}

For maintainers

If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.

[jev on Agent Plugins Marketplace](https://pluginsmp.com/plugins/jev-3)