Jev MCP
Use before a step that judges material you already have — a claim against evidence, screening fetched text before reading it, classifying or keeping or dropping many items, finding or reranking candidates, one bounded choice, comparing two passages, grading on an ordered rubric, judging a file without reading it into your context, extracting a regex-matched value, reviewing a patch, or checking that work finished. Batch every question about one state into one call. Jev is invoked when an unresolved judgment earns a model decision. Deterministic evidence takes precedence; Jev is not a mandatory
Install
npx skills add PyModel/jev-judge-mcp --skill jev-mcpgit clone --depth 1 https://github.com/PyModel/jev-judge-mcp.git /tmp/jev-judge-mcp
mkdir -p ~/.claude/skills
cp -r /tmp/jev-judge-mcp/src/jev_judge_mcp/skills/jev-mcp ~/.claude/skills/jev-mcpPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Routing a judgment This skill routes this server's tools. Jev is invoked when an unresolved judgment earns a model decision. Deterministic evidence takes precedence; Jev is not a mandatory ceremony. High-value calls: before a done claim, `jev_gate`, unless tests, type checks, build, lint, or another explicit acceptance criterion already settle completion; before reading fetched or pasted external text, `jev_screen`; checking another agent's report or research claims, `jev_verify`. Skip it when the answer is already determined by a test, type-check, or the code itself; when the choice is trivial or cheap to reverse; when the question cannot be enumerated into bounded options; or when the same unchanged decision was already asked. Resource `jev-skill://jev/SKILL.md` (prompt `jev`) is for building an app that calls the Jev API. It does not say which tool to call. Do not copy its cookbook thresholds onto these tools; this server's defaults are its own (`docs/reference/limits.md`). The same on-demand rule applies there. Jev answers a typed Question about a State and returns probabilities. Policy turns the validated answer into an Action. A Verdict is the semantic conclusion for one item (verified, contradicted, unsupported, and the other per-tool words). The Action is what you may do with that Verdict. Those words are defined in `docs/CONTEXT.md`.
Details
- Skill name
jev-mcp- Repository
- PyModel/jev-judge-mcp
- Path
src/jev_judge_mcp/skills/jev-mcp/SKILL.md- License
- MIT
- Repo updated
- 26 October 2026
FAQ
What does the Jev MCP skill do?
Use before a step that judges material you already have — a claim against evidence, screening fetched text before reading it, classifying or keeping or dropping many items, finding or reranking candidates, one bounded choice, comparing two passages, grading on an ordered rubric, judging a file without reading it into your context, extracting a regex-matched value, reviewing a patch, or checking that work finished. Batch every question about one state into one call. Jev is invoked when an unresolved judgment earns a model decision. Deterministic evidence takes precedence; Jev is not a mandatory
How do I install the Jev MCP skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add PyModel/jev-judge-mcp --skill jev-mcp. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Jev MCP popular?
It lives in PyModel/jev-judge-mcp, which has 49 GitHub stars and 5 forks.
Related skills
Jev
49PyModel/jev-judge-mcp
Use when code needs a judgment about natural language that rules or regex cannot make (classify, route, triage, moderate, score, rank, match, dedupe, filter, extract, verify, or gate an action), when brainstorming where AI could fit in an app, when an LLM call exists only to pick a label, score, or yes/no, or when code uses TypeSafe AI, Jev, System One, typesafe-sdk, or @typesafe-ai/sdk.
Test Audit
49PyModel/jev-judge-mcp
Invoke whenever writing, changing, reviewing, or sweeping tests in jev-judge-mcp. Authoring gate for new tests plus audit workflow for low-value, implementation-coupled, or duplicative tests and the test-only production seams they demand, under this repo's rule that a test is never deleted or weakened to make a failing build pass.
Lark MCP
37whatevertogo/FeiShuSkill
飞书/Lark 官方 MCP 集成。支持发送消息、创建群组、操作多维表格(Bitable)、导入/搜索文档、知识库查询。触发词:飞书、Feishu、Lark、多维表格、bitable、飞书文档、飞书群。
Agent Builder
49vitoworleone/claude-code-handbook
Design and build AI agents for any domain. Use when users: (1) ask to "create an agent", "build an assistant", or "design an AI system" (2) want to understand agent architecture, agentic patterns, or autonomous AI (3) need help with capabilities, subagents, planning, or skill mechanisms (4) ask about Claude Code, Cursor, or similar agent internals (5) want to build agents for business, research, creative, or operational tasks Keywords: agent, assistant, autonomous, workflow, tool use, multi-step, orchestration
Codex Claude Collab
49ashinno/Codex-Claude-Collab
Coordinate Codex with Claude Code in real time for collaborative software work. Use when the user wants Codex to plan or review while Claude Sonnet or Opus implements, when creative/frontend/product work should be delegated to Claude, when a Codex-to-Claude handoff loop is requested, or when phrases like "notify Sonnet", "notify Opus", "Claude gets the work done", "Codex reviews", "realtime Claude Code", or "Claude Code delegation" appear.
Booking Best Price
34aiwithremy/claude-skills-booking-best-price
Use when the user pastes a holiday-rental or hotel listing link (Airbnb, Stayz, Vrbo, Expedia, Booking.com, HomeToGo, or similar) and wants to identify the actual property, find it listed on other sites, compare prices to book it cheaper, or track down the property manager / owner contact to book direct. Requires the Firecrawl MCP server. This skill verifies Firecrawl is installed and walks the user through setup first if it is not.
Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.