New: 97,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Hyper Jev

by disler in disler/ten-levels-of-jev

Integrate and use Jev, TypeSafe AI's System One decision model, in production codebases.

MITAuto-check passedAI & LLM Engineering

Install Hyper Jev

skills CLI
$ npx skills add disler/ten-levels-of-jev --skill hyper-jev -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install disler/ten-levels-of-jev hyper-jev --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/disler/ten-levels-of-jev.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/hyper-jev .claude/skills/hyper-jev && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
hyper-jev
GitHub stars
202
Token cost
~1.7k tokens
SKILL.md length
826 words
Files
105
Skills in repo
1
Repo updated
First seen
Licence
MIT

At a glance

Integrate and use Jev, TypeSafe AI's System One decision model, in production codebases.

  • Works in 8 steps: Identify the decision. State the input,… → Read the relevant cookbook. Start with… → Place the code. For a new service, copy… → …
  • TypeSafe/OpenRouter decision APIs
  • SKILL.md covers Purpose, Instructions, Workflow and Examples, plus 1 more section
  • Needs TYPESAFE_API_KEY and OPENROUTER_API_KEY

What it does

Hyper Jev is an agent skill from disler/ten-levels-of-jev. Integrate and use Jev, TypeSafe AI's System One decision model, in production codebases. Use for hyper-jev, Jev setup or deployment, TypeSafe/OpenRouter decision APIs, noul/choice/score questions, classifiers, routing, confidence gates, guardrails, dynamic options, retries, cost tracking, and raw payload retention. Includes a portable TypeScript client, tested examples, and a Jev cookbook. Not a presentation or UI-generation skill.

Its SKILL.md is about 1.7k tokens, which your agent loads only when the skill is triggered. The skill folder holds 106 other files (for example `cookbook/01-structure-and-setup.md`, `cookbook/02-input-output.md` and `cookbook/03-question-design.md`).

It sits in AI & LLM Engineering, covering Model routing and gateways and LLM cost and token optimization. It works with OpenRouter and TypeScript. The repository describes itself as: Ten levels of Jev, from one smart if statement to a coding agent that reaches for Jev on its own. The licence is MIT.

When your agent uses it

  • TypeSafe/OpenRouter decision APIs
  • Noul/choice/score questions
  • Confidence gates
  • Dynamic options

Example prompts

  • “/hyper-jev”

Requirements

  • A credential in TYPESAFE_API_KEY
  • A credential in OPENROUTER_API_KEY

Workflow steps

8 steps, taken from the first numbered list in SKILL.md.

  1. Identify the decision. State the input, allowed outcomes, uncertainty path, and any side effects. Choose Noul for a yes/no probability…
  2. Read the relevant cookbook. Start with the index. Read structure and setup, input/output, and client operations for a new integration. Use…
  3. Place the code. For a new service, copy the standalone starter into a new directory. In an existing app, adapt src/core/ to its backend…
  4. Initialize once. Load environment configuration before construction. Create one shared client per process or dependency-injection scope…
  5. Define the contract. Minimize state. Describe each criterion as an observable situation. Give open-ended choices an other outcome. Keep…
  6. Apply policy in code. Narrow answer types. Distinguish selected class from confidence. Preserve an explicit review path for ambiguity or…
  7. Test offline first. Run the bundled tests and new boundary tests. Use mocked HTTP responses to test provider selection, retries, malformed…
  8. Prepare for production. Calibrate on labeled domain examples. Review credentials, request size, deadlines, concurrency, budget, audit…

What it can do on your machine

Read from SKILL.md and the folder at commit 777adaf. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names these keys or tokens, usually read from environment variables:

    • TYPESAFE_API_KEY
    • OPENROUTER_API_KEY

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Hyper Jev loads about 1.7k tokens when it runs. Until then it costs about 111 tokens; SKILL.md has 826 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~111
When it runs · the whole SKILL.md, loaded when a task matches
~1.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from disler/ten-levels-of-jev at commit 777adaf, republished under its MIT licence (© disler). 826 words, ~1,691 tokens.

Download SKILL.mdSave it as .claude/skills/hyper-jev/SKILL.md (or your agent's skills folder). This skill also uses 104 other files; get the full folder from GitHub.
name
hyper-jev
description
Integrate and use Jev, TypeSafe AI's System One decision model, in production codebases. Use for hyper-jev, Jev setup or deployment, TypeSafe/OpenRouter decision APIs, noul/choice/score questions, classifiers, routing, confidence gates, guardrails, dynamic options, retries, cost tracking, and raw payload retention. Includes a portable TypeScript client, tested examples, and a Jev cookbook. Not a presentation or UI-generation skill.
argument-hint
request prompt

Hyper Jev

Purpose

Turn the user's request into a small, tested Jev integration. Jev evaluates a state against bounded questions and returns typed answers. Your application owns calculations, permissions, thresholds, and execution.

This skill packages the Jev core and thirty examples from the ten-levels codebase. Deploy means integrate the hosted API into the user's backend, not self-host model weights. No slide decks, web lab, visualizations, or agent runtime are required.

Instructions

  • Treat $ARGUMENTS as the user's request prompt. Infer the task from the conversation if arguments are absent. Ask only for a missing detail that changes the implementation.
  • Resolve bundled paths relative to this SKILL.md, not the target project's working directory. The entire skill is portable and lives in .claude/skills/hyper-jev/.
  • For explanation-only requests, answer without changing files. For implementation, inspect the destination backend, package manager, configuration, tests, and existing conventions before editing.
  • Reuse the bundled client, types, builders, and their sibling dependencies. Do not invent a chat-completions adapter or a second Jev transport.
  • Select the provider once per client instance. Prefer TYPESAFE_API_KEY, otherwise OPENROUTER_API_KEY. Explicit provider overrides are supported. Never switch providers because of a failed request. Missing credentials must not silently turn production decisions into mock results.
  • Keep keys on the server. Do not print credentials, include Authorization headers in audit records, or send sensitive state to a provider without an approved data policy.
  • Preserve the complete client result through the service boundary. Business decisions are derived views, not replacements for the raw request, raw response, usage, model ID, and metadata.
  • Keep each use case's questions, thresholds, and pure policy function together. Ask independent questions in one call. Make another request only when its state or choices depend on the preceding answer.
  • Use Jev for bounded judgments, not arithmetic, date comparisons, unrestricted text generation, or authorization. Security gates are an additional signal, never the only control.
  • Copy only the runtime core and requested use cases into an existing app. The thirty-example starter is a reference, not a requirement to add every example.
  • Keep changes at the requested scope. Do not deploy infrastructure, make paid calls, modify unrelated modules, or commit unless requested.

Workflow

  1. Identify the decision. State the input, allowed outcomes, uncertainty path, and any side effects. Choose Noul for a yes/no probability, Choice for a finite label, or Score for an ordered rubric.
  2. Read the relevant cookbook. Start with the index. Read structure and setup, input/output, and client operations for a new integration. Use question design, production checks, and only the needed use-case guides.
  3. Place the code. For a new service, copy the standalone starter into a new directory. In an existing app, adapt src/core/ to its backend integration folder and add a named decision module. Preserve the client's provider, retry, contract, raw-payload, and cost behavior.
  4. Initialize once. Load environment configuration before construction. Create one shared client per process or dependency-injection scope. Verify the selected provider name, not the key. Use an explicit mock only in development and tests. Construct a new client to rotate a key.
  5. Define the contract. Minimize state. Describe each criterion as an observable situation. Give open-ended choices an other outcome. Keep option IDs stable and map them to allowed code paths. Validate untrusted requests before sending.
  6. Apply policy in code. Narrow answer types. Distinguish selected class from confidence. Preserve an explicit review path for ambiguity or outages. Retain the result alongside the decision, following the starter service.
  7. Test offline first. Run the bundled tests and new boundary tests. Use mocked HTTP responses to test provider selection, retries, malformed responses, raw retention, and accounting without sending data or spending money. The deterministic mock tests wiring, not model accuracy.
  8. Prepare for production. Calibrate on labeled domain examples. Review credentials, request size, deadlines, concurrency, budget, audit retention, and side-effect permissions. Run a real API smoke test only when authorized, then report the provider and resolved model actually observed.
Show full SKILL.md (176 more words)Show less

Examples

/hyper-jev add Jev support triage to our Node API

Inspect the API service. Reuse the client, fan out category/blocked/frustration questions, add a review outcome, return { decision, result }, and test routes separately from transport.

/hyper-jev explain noul vs choice vs score with inputs and outputs

Read the input/output and question-design guides. Show one concrete payload and explain which decisions belong in code. Do not scaffold files.

/hyper-jev gate agent tool calls and track their cost

Use the guardrails guide. Combine deterministic tool permissions with a risk classification and review path. Keep all probabilities and wire bodies. Separate reported charges, estimates, and unknown cost.

/hyper-jev classify documents across 2,000 categories

Use the high-cardinality guide. Prune or traverse a taxonomy rather than exceeding 255 choices. Bound depth, requests, and spend. Preserve candidates' real IDs and treat path scores as ranking signals.

Report Format

Report the implemented decision and file paths, selected provider behavior, tests actually run, and remaining production checks. Distinguish offline contract tests from live validation. Do not describe a mock run as a successful live deployment.

© disler, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 104 other files in .claude/skills/hyper-jev of disler/ten-levels-of-jev.

  • SKILL.md
  • cookbook/01-structure-and-setup.md
  • cookbook/02-input-output.md
  • cookbook/03-question-design.md
  • cookbook/04-client-operations.md
  • cookbook/05-production.md
  • cookbook/README.md
  • cookbook/use-cases/01-single-decisions.md
  • cookbook/use-cases/02-fan-out.md
  • cookbook/use-cases/03-composite-scoring.md
  • cookbook/use-cases/04-confidence-gating.md
  • cookbook/use-cases/05-routing.md
  • cookbook/use-cases/06-interactive-decisions.md
  • cookbook/use-cases/07-guardrails.md
  • cookbook/use-cases/08-decision-loops.md
  • cookbook/use-cases/09-high-cardinality.md
  • cookbook/use-cases/10-dynamic-questions.md
  • cookbook/use-cases/README.md
  • templates
  • … and 86 more

Open the folder on GitHubat commit 777adaf

Compare with similar skills

Hyper Jev next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Hyper Jev compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Hyper Jev this skilldisler/ten-levels-of-jev202—~1.7kAutomated safety check: PassMIT
Pinme LLMglitternetwork/pinme3.7k1 repos~2.8kAutomated safety check: PassMIT
FreeRide Free Model ManagerShaivpidadi/FreeRide2383 repos~1.1kAutomated safety check: PassNone
Olore Openrouter Latestolorehq/olore103—~579Automated safety check: PassMIT
Jev Model Routingkerpopule/hermes-jev-skills1k—~2.7kAutomated safety check: PassMIT
Hot Monitorliyupi/yupi-hot-monitor7171 repos~1.2kAutomated safety check: PassNone

Similar skills

  • Pinme LLM

    glitternetwork/pinme

    A skill your agent uses when a PinMe project (Worker TypeScript) needs to call OpenRouter-backed LLM APIs, including models, chat/completions, streaming, or OpenRouter web search.

    3.7k GitHub starsUsed in 1 repo~2.8k tokens
    AI & LLM EngineeringAuto-check passed

  • FreeRide Free Model Manager

    Shaivpidadi/FreeRide

    Configures OpenClaw to use free OpenRouter models, setting the best one as primary and adding ranked fallbacks so rate limits do not interrupt work.

    238 GitHub starsUsed in 3 repos~1.1k tokens
    AI & LLM EngineeringAuto-check passed

  • Olore Openrouter Latest

    olorehq/olore

    Local OpenRouter documentation reference (latest). An agent skill from olorehq/olore.

    103 GitHub stars~579 tokensUpdated today
    AI & LLM EngineeringAuto-check passed

  • Jev Model Routing

    kerpopule/hermes-jev-skills

    Routes a turn or delegated task to the cheapest model and effort lane that will still do it right, using the Jev decision model to classify difficulty and escalate only when needed.

    1k GitHub stars~2.7k tokensUpdated yesterday
    AI & LLM EngineeringAuto-check passed

  • Hot Monitor

    liyupi/yupi-hot-monitor

    AI hotspot monitoring and trending topic discovery across multiple sources (Bing, Google, DuckDuckGo, HackerNews, Sogou, Bilibili, Weibo, Twitter).

    717 GitHub starsUsed in 1 repo~1.2k tokens
    DatabasesAuto-check passed

  • Embeddings via 9Router

    decolua/9router

    Generates vector embeddings through the 9Router /v1/embeddings endpoint, using models from providers such as OpenAI, Gemini, Mistral and Voyage for RAG and semantic search.

    30k GitHub stars~604 tokensUpdated 6 days ago
    AI & LLM EngineeringAuto-check passed

Works with

Categories

Questions about Hyper Jev

What does Hyper Jev do?

Integrate and use Jev, TypeSafe AI's System One decision model, in production codebases. Hyper Jev is an agent skill from disler/ten-levels-of-jev. Integrate and use Jev, TypeSafe AI's System One decision model, in production codebases.

When should I use Hyper Jev?

Hyper Jev fits situations like: typeSafe/OpenRouter decision APIs; noul/choice/score questions; confidence gates; dynamic options.

How do I install Hyper Jev in Claude Code?

Run `npx skills add disler/ten-levels-of-jev --skill hyper-jev -a claude-code`. Or copy the skill folder (.claude/skills/hyper-jev in disler/ten-levels-of-jev) into .claude/skills/hyper-jev in your project. Claude Code loads it when a task matches its description.

How do I install Hyper Jev in Codex?

Run `npx skills add disler/ten-levels-of-jev --skill hyper-jev -a codex`. Or copy the skill folder (.claude/skills/hyper-jev in disler/ten-levels-of-jev) into .agents/skills/hyper-jev in your project. Codex loads it when a task matches its description.

Can I use Hyper Jev in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add disler/ten-levels-of-jev --skill hyper-jev -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/hyper-jev, .gemini/skills/hyper-jev, .github/skills/hyper-jev and .opencode/skills/hyper-jev in your project.

What does Hyper Jev need to run?

Going by SKILL.md and its folder, Hyper Jev needs credentials named TYPESAFE_API_KEY and OPENROUTER_API_KEY. Our summary lists: A credential in TYPESAFE_API_KEY; A credential in OPENROUTER_API_KEY.

Does Hyper Jev access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Hyper Jev safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Hyper Jev use?

Hyper Jev is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Hyper Jev use?

About 1.7k tokens (SKILL.md is roughly 6.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Hyper Jev?

Skills that share tags, products or a category with Hyper Jev: Pinme LLM (glitternetwork/pinme, 3.7k stars), FreeRide Free Model Manager (Shaivpidadi/FreeRide, 238 stars), Olore Openrouter Latest (olorehq/olore, 103 stars) and Jev Model Routing (kerpopule/hermes-jev-skills, 1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Hyper Jev?

disler (a GitHub user) maintains it in disler/ten-levels-of-jev, which has 202 GitHub stars. The repository was last updated on September 27, 2026.

Source: disler/ten-levels-of-jev on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.