COMMUNITY OBSERVEDMetric claims: AUTHOR REPORTEDSOURCE AUDITED⚡ Jev Skills

so_ainsight Jev Skills Workflow

A collection of Japanese open-source projects covering context compaction, browser agents, search and more.

so_ainsight Jev Skills Workflow preview from the original source
Show the original post (Japanese)

超高速AI「Jev」を活用した注目のオープンソース8選:

1. fast-jev-compaction

Claude Codeの要約待ちを解消。不要なツール履歴だけを瞬時に削り、会話の文脈を保ちます。

https://github.com/tamaratran/fast-jev-compaction

2. jev-browser

LLMが全体方針を立て、Jevが画面内のボタン選択やクリックを実行。MCPにも対応。

https://github.com/Ying-Kai-Liao/jev-browser

3. jev-search

Web検索を高速化。検索文の意図理解から情報源の選択、結果の絞り込みまでJevが担当。

https://github.com/superagents-lab/jev-search

4. Reticle

AIエージェントに画面の知覚能力を追加。操作しているWebやアプリの画面構造を理解。

https://github.com/reticlehq/reticle

5. NanoJev

Jevの仕組みを軽量に再現したプロジェクト。並列判断や動的候補、学習の流れを学べます。

https://github.com/TianyuCodings/NanoJev

6. open-alternative-jev

オープンなAIモデルを使ってJev風の高速判断レイヤーを構築。自前環境で動かせます。

https://github.com/ikermoel/open-alternative-jev

7. pi-jev

PiコーディングエージェントにJevを接続。ツールの実行制御と型付きの確実な出力を実現。

https://github.com/y0usaf/pi-jev

8. hermes-jev-skills

エージェントにJevを組み込み、モデルの切り替えや記憶管理、スキルの自動選択を担当。

https://github.com/kerpopule/hermes-jev-skills

— そう|Claude CodeのX運用で月800万インプ

How Jev fits in the loop

  1. Agent collects runtime context or input state from external sources.
  2. Bounded decision options are formatted into Jev System One questions.
  3. Jev scores each candidate option and returns a probability for it.
  4. Deterministic code executes the selected action or routes to specialized models.

How to reproduce

  1. Read the author's original post on X at https://x.com/so_ainsight/status/2102173541135466976.
  2. Inspect the described workflow architecture and bounded decision points.
  3. Test Jev System One decision queries against the provided task context.

Why this build matters

Demonstrates real-world Jev Skills experimentation by @so_ainsight with verified community reach of 37.2k views.

Limitations

  • Implementation details are based on the author's public post on X.
  • Performance figures and benchmarks are author-reported community claims.
  • Production deployments may require custom calibration and policy thresholds.

Patterns