jev-style plugin
Skills (serve, use, eval, adopt, guard, mcp) and MCP tools (decide, noul, choice, score) for the local Jev-Style decision model. Needs `jev-style serve` running.
How to install the jev-style plugin
Run both commands inside Claude Code. The first adds the marketplace once; the second installs the plugin.
/plugin marketplace add lawrence3699/jev-style
/plugin install jev-style@jev-styleWhat's inside jev-style
6 parts read from the plugin's own files.
Skills 6
Claude loads it when the task matches- jev-style
Call the local Jev-Style decision model (systemone-compatible API, default http://127.0.0.1:8765) to answer typed questions about text or JSON - yes/no probabilities (noul), one-of-N choices, ordered scores - for routing, classification, gating, triage, guardrails and scoring. Use when writing code that classifies, routes, scores or checks text with a small local model, when replacing an LLM call that only returns a label, or when the user mentions jev-style, /v1/systemone, noul, choice or score questions.
- jev-style-adopt
Find LLM API calls in a codebase that only return a decision (a category, yes/no, or a 1-5 rating) and move them to the local Jev-Style decision model safely - scan, pick candidates with the user, rewrite each as a typed noul/choice/score question with a confidence cascade back to the LLM, run in shadow mode, measure agreement with jev-style eval, then switch. Use when the user wants to cut LLM cost or latency for classification, routing, moderation, gating or scoring calls, or asks which LLM calls a small local model could replace.
- jev-style-eval
Measure the Jev-Style local decision model on the user's own labelled examples before relying on it - accuracy, Brier, ECE, how many decisions can be automated at a 1/5/10 % error budget and at which confidence threshold, plus a refitted calibration temperature. Use when the user asks whether the local model is good enough for their task, wants thresholds for automating a classification/routing/gating step, or wants to compare it with the labels an existing LLM or human process produces.
- jev-style-guard
Install the Jev-Style guard, a Claude Code PreToolUse hook that has a local decision model check every Bash, Write, Edit, Read, WebFetch and MCP tool call for destructive actions, data exfiltration, secrets access and out-of-project paths, and answers allow / ask / deny. Use when the user wants a local safety net or approval gate for Claude Code (or another agent with PreToolUse-style hooks), wants to block recursive deletes, force-pushes or key exfiltration, or wants to tune or measure such a guard.
- jev-style-mcp
Connect the Jev-Style local decision model to an AI coding agent as an MCP server (tools decide, noul, choice, score, model_info) - Claude Code, Claude Desktop, Cursor, Codex CLI, Windsurf or any MCP client. Use when the user wants their agent to classify, route, gate or score text with a local model through MCP, or asks to add jev-style as an MCP server.
- jev-style-serve
Install and start the Jev-Style local decision server (systemone-compatible API + Playground at http://127.0.0.1:8765) on macOS, Linux or Windows/WSL, pick the right backend (MLX on Apple silicon, PyTorch CUDA/CPU elsewhere, llama.cpp GGUF optional), verify it answers, and optionally keep it running at login. Use when the user wants to run Jev-Style, set up a local decision model / systemone endpoint, or when another jev-style skill finds no server on :8765.
Using jev-style in Claude Code
Skills need no command. Claude reads them when your request matches what they describe.
Update, turn off or remove
- Update the plugin (shell)
- claude plugin update jev-style@jev-style
- Refresh the marketplace
- /plugin marketplace update jev-style
- Turn it off for now
- /plugin disable jev-style@jev-style
- Remove it
- /plugin uninstall jev-style@jev-style
Skills in jev-style
- jev-stylelawrence3699/jev-style
Call the local Jev-Style decision model (systemone-compatible API, default http://127.0.0.1:8765) to answer typed questions about text or JSON - yes/no probabilities (noul), one-of-N choices, ordered scores - for routing, classification, gating, triage, guardrails and scoring. Use when writing code that classifies, routes, scores or checks text with a small local model, when replacing an LLM call that only returns a label, or when the user mentions jev-style, /v1/systemone, noul, choice or score questions.
2 files135stars - jev-style-adoptlawrence3699/jev-style
Find LLM API calls in a codebase that only return a decision (a category, yes/no, or a 1-5 rating) and move them to the local Jev-Style decision model safely - scan, pick candidates with the user, rewrite each as a typed noul/choice/score question with a confidence cascade back to the LLM, run in shadow mode, measure agreement with jev-style eval, then switch. Use when the user wants to cut LLM cost or latency for classification, routing, moderation, gating or scoring calls, or asks which LLM calls a small local model could replace.
2 files135stars - jev-style-evallawrence3699/jev-style
Measure the Jev-Style local decision model on the user's own labelled examples before relying on it - accuracy, Brier, ECE, how many decisions can be automated at a 1/5/10 % error budget and at which confidence threshold, plus a refitted calibration temperature. Use when the user asks whether the local model is good enough for their task, wants thresholds for automating a classification/routing/gating step, or wants to compare it with the labels an existing LLM or human process produces.
135stars - jev-style-guardlawrence3699/jev-style
Install the Jev-Style guard, a Claude Code PreToolUse hook that has a local decision model check every Bash, Write, Edit, Read, WebFetch and MCP tool call for destructive actions, data exfiltration, secrets access and out-of-project paths, and answers allow / ask / deny. Use when the user wants a local safety net or approval gate for Claude Code (or another agent with PreToolUse-style hooks), wants to block recursive deletes, force-pushes or key exfiltration, or wants to tune or measure such a guard.
135stars - jev-style-mcplawrence3699/jev-style
Connect the Jev-Style local decision model to an AI coding agent as an MCP server (tools decide, noul, choice, score, model_info) - Claude Code, Claude Desktop, Cursor, Codex CLI, Windsurf or any MCP client. Use when the user wants their agent to classify, route, gate or score text with a local model through MCP, or asks to add jev-style as an MCP server.
135stars - jev-style-servelawrence3699/jev-style
Install and start the Jev-Style local decision server (systemone-compatible API + Playground at http://127.0.0.1:8765) on macOS, Linux or Windows/WSL, pick the right backend (MLX on Apple silicon, PyTorch CUDA/CPU elsewhere, llama.cpp GGUF optional), verify it answers, and optionally keep it running at login. Use when the user wants to run Jev-Style, set up a local decision model / systemone endpoint, or when another jev-style skill finds no server on :8765.
135stars
jev-style plugin questions
How do I install the jev-style plugin?
In Claude Code run /plugin marketplace add lawrence3699/jev-style, then /plugin install jev-style@jev-style.
What does the jev-style plugin include?
jev-style bundles 6 skills.
How do I update or remove the jev-style plugin?
Update it with claude plugin update jev-style@jev-style in your shell, or from the Installed tab of /plugin. Remove it with /plugin uninstall jev-style@jev-style.
What license is the jev-style plugin under?
Apache-2.0, as stated in the plugin's manifest.