Skip to content
Back to search

optim-jev

MarketplaceWorks with Claude CodeGitHub

Claude Code optimization tools from oqura.ai.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked yesterday
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/oqura-ai--optim-jev/badge.svg)](/listings/oqura-ai--optim-jev)

HTML

<a href="/listings/oqura-ai--optim-jev"><img src="/api/listings/oqura-ai--optim-jev/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Requirements

- [Claude Code](https://code.claude.com/docs/en/overview) with plugin support - Python 3.10 or newer - A TypeSafe API key for Jev

Install

git clone https://github.com/Oqura-ai/optim-jev.git
claude plugin marketplace add ./optim-jev
claude plugin install optim-jev@optim-jev

Set the key in Claude Code under `/config` → **TypeSafe API key**, or export it before starting Claude Code:

export TYPESAFE_API_KEY="your-key"

Restart Claude Code. jev-compacter is ready immediately; enable the router and skill picker when you want them. See [Getting started](docs/getting-started.md) for every command, option, and troubleshooting step.

> [!NOTE] > The plugin uses only Python's standard library. Your API key is passed to child processes through the environment and is never written to project files or logs.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 17 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. {
  2. "name": "optim-jev",
  3. "description": "Claude Code optimization tools from oqura.ai.",
  4. "owner": {
  5. "name": "oqura.ai",
  6. "url": "https://github.com/Oqura-ai"
  7. },
  8. "plugins": [
  9. {
  10. "name": "optim-jev",
  11. "source": "./",
  12. "description": "Token-usage optimization tools for Claude Code workflows.",
  13. "version": "0.1.0",
  14. "license": "MIT"
  15. }
  16. ]
  17. }
README.md · README · 184 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. <div align="center">
  2. # optim-jev
  3. **Spend fewer tokens in long agent sessions without changing how you work.**
  4. Jev-powered optimization for Claude Code, plus model routing for OpenAI Codex.
  5. [![Claude Code plugin](https://img.shields.io/badge/Claude_Code-plugin-D97757?style=flat-square)](https://code.claude.com/docs/en/overview)
  6. [![Codex plugin](https://img.shields.io/badge/OpenAI_Codex-plugin-111827?style=flat-square)](https://developers.openai.com/codex/plugins)
  7. [![License: MIT](https://img.shields.io/badge/License-MIT-22C55E?style=flat-square)](#license)
  8. [![Our Discord](https://img.shields.io/badge/Discord-Join%20Community-5865F2?style=flat-square&logo=discord&logoColor=white)](https://discord.gg/EpUxGjNBva)
  9. [Get started](#claude-code-quick-start) · [Codex support](docs/codex.md) · [Documentation](docs/README.md) · [Architecture](docs/architecture.md) · [Contributing](docs/contributing.md)
  10. <img src="assets/demo.gif" alt="optim-jev demo in Claude Code" width="100%">
  11. </div>
  12. ## Why optim-jev?
  13. <img src="assets/optim-jev.png" alt="optim-jev architecture: context compaction, model routing, and skill selection" width="100%">
  14. Long agent sessions accumulate stale tool output, send oversized skill catalogs, and use the main model for work a smaller model could handle. Claude Code supports all three optimizers below. Codex supports the model router and an off-by-default advisory skill experiment.
  15. Each tool works independently and fails open if unavailable. Measured cost reports remain Claude-only.
  16. ## Three tools, one project
  17. | Tool | What it does | Available in | Claude default | Learn more |
  18. |---|---|---|:---:|---|
  19. | **jev-compacter** | Removes stale tool calls and outputs instead of summarizing away the whole conversation | Claude Code | On | [Guide](docs/jev-compacter.md) |
  20. | **jev-router** | Runs each subagent, Explore searches included, on the cheapest model that fits, within project rules that evolve with your workflow | Claude Code and Codex | Off | [Guide](docs/jev-router.md) |
  21. | **jev-skills** | Sends a short, relevant skill list for each prompt instead of the entire catalog | Claude Code; experimental advisory mode in Codex | Off | [Guide](docs/jev-skills.md) |
  22. <details>
  23. <summary><strong>jev-compacter — preserve useful context</strong></summary>
  24. <br>
  25. <img src="assets/jev-compacter.png" alt="jev-compacter session statistics" width="100%">
  26. It works automatically with `/compact` and auto-compaction. Inspect the current session with:
  27. ```text
  28. /jev-compact stats
  29. ```
  30. Want it to prune earlier or keep more? Pick a **Compaction: style** (`balanced`, `save-tokens`, or `keep-context`) in `/config` under optim-jev. See [Pick a style](docs/jev-compacter.md#pick-a-style).
  31. </details>
  32. <details>
  33. <summary><strong>jev-router — the right model for every subagent</strong></summary>
  34. <br>
  35. <img src="assets/jev-router.png" alt="jev-router model routing decisions" width="100%">
  36. Your main conversation keeps its model, so its prompt cache stays warm. The routing happens where it pays off: every subagent Claude spawns gets its own model and effort from Jev, Explore searches included. A quick lookup runs on haiku; a deep architecture review gets a stronger model.
  37. Claude works as usual, editing directly and delegating when a task calls for it or when you ask. `rules.json` sets which models may edit each path; when a path needs a different model than your session's, Claude hands that change to a routed subagent.
  38. ```text
  39. /jev-route init "what matters in this project"
  40. /jev-route on | off
  41. /jev-route why | stats
  42. ```
  43. Tell Claude to change a preference later—for example, “let haiku edit the tests”—and the project rules update with your workflow.
  44. </details>
  45. <details>
  46. <summary><strong>jev-skills — load only relevant skills</strong></summary>
  47. <br>
  48. <img src="assets/jev-skills.png" alt="jev-skills prompt shortlist" width="100%">
  49. ```text
  50. /jev-skills on | off
  51. /jev-skills scope project | global
  52. /jev-skills why | stats | reindex
  53. ```
  54. </details>
  55. ## Claude Code quick start
  56. ### Requirements
  57. - [Claude Code](https://code.claude.com/docs/en/overview) with plugin support
  58. - Python 3.10 or newer
  59. - A TypeSafe API key for Jev
  60. ### Install
  61. ```bash
  62. git clone https://github.com/Oqura-ai/optim-jev.git
  63. claude plugin marketplace add ./optim-jev
  64. claude plugin install optim-jev@optim-jev
  65. ```
  66. Set the key in Claude Code under `/config` → **TypeSafe API key**, or export it before starting Claude Code:
  67. ```bash
  68. export TYPESAFE_API_KEY="your-key"
  69. ```
  70. Restart Claude Code. jev-compacter is ready immediately; enable the router and skill picker when you want them. See [Getting started](docs/getting-started.md) for every command, option, and troubleshooting step.
  71. > [!NOTE]
  72. > The plugin uses only Python's standard library. Your API key is passed to child processes through the environment and is never written to project files or logs.
  73. ## OpenAI Codex support — new
  74. Start with the complete [Codex guide](docs/codex.md) for installation, usage, model-tier setup,
  75. verification, limitations, troubleshooting, and uninstall steps.
  76. optim-jev can automatically route each Codex subagent to an account-specific `fast`, `balanced`, or
  77. `strong` model. It also selects reasoning effort, applies project path rules, escalates after
  78. failures, and can provide an optional advisory skill shortlist. Claude and Codex keep separate
  79. model catalogues, policies, caches, and logs.
  80. > [!IMPORTANT]
  81. > **Compaction is not supported in Codex.** Codex hooks cannot replace live conversation history.
  82. > The AbovePrompt dashboard, slash commands, and measured token and cost reports are also
  83. > Claude-only.
  84. ### Install and use
  85. Codex CLI 0.162.1 or newer can install optim-jev directly from GitHub. No PyPI release or manual
  86. clone is required.
  87. ```powershell
  88. py -3 -m pip install "https://github.com/Oqura-ai/optim-jev/archive/refs/heads/main.zip"
  89. optim-jev codex install
  90. $env:TYPESAFE_API_KEY = "your-key"
  91. cd C:\path\to\your-project
  92. optim-jev codex setup
  93. optim-jev codex status
  94. codex
  95. ```
  96. On the first Codex launch after install or update, run `/hooks`, review and trust every optim-jev
  97. hook, then restart Codex.
  98. Inside Codex, subagent routing is automatic. Check the most recent decision or control project
  99. features with:
  100. ```text
  101. !optim-jev codex why
  102. !optim-jev codex status
  103. !optim-jev codex on
  104. !optim-jev codex off
  105. !optim-jev codex skills-on
  106. !optim-jev codex skills-off
  107. ```
  108. ## Documentation
  109. | Resource | What you will find |
  110. |---|---|
  111. | [Getting started](docs/getting-started.md) | Installation, configuration, commands, and troubleshooting |
  112. | [Codex support](docs/codex.md) | Codex installation, router setup, configuration, and limitations |
  113. | [TinyCart demo](demo/tinycart/DEMO.md) | A short, repeatable showcase of all three tools on a toy project |
  114. | [Performance metrics](docs/performance-metrics.md) | How token usage, cost, and estimated savings are calculated |
  115. | [Architecture](docs/architecture.md) | The TypeScript-to-Python bridge, data flow, and session logs |
  116. | [Contributing](docs/contributing.md) | Local setup, tests, conventions, and pull requests |
  117. ## Community and support
  118. - Found a bug or have a feature idea? [Open an issue](https://github.com/Oqura-ai/optim-jev/issues/new/choose).
  119. - Want to help? Read the [contributing guide](docs/contributing.md) and submit a focused pull request.
  120. - For the wider Claude Code community, join the official [Claude Developers Discord](https://anthropic.com/discord).
  121. ## License
  122. optim-jev is released under the MIT License.
  123. <div align="center">
  124. Built by [Oqura.ai](https://github.com/Oqura-ai) for more efficient coding-agent workflows.
  125. </div>

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.