How do I install the Jev Act skill?

Run npx skillmds@latest add wuyoscar/jev-act in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Jev Act safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: makes network calls. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Jev Act?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Jev Act free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Jev Act?

wuyoscar (@wuyoscar) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Jev Act by wuyoscar — Agent Skill

Name: Jev Act
Rating: 4.2 (5 reviews)
Author: wuyoscar

File contents

Choose the next action

Use safely

Choose the service once and keep that choice. If unset, ask A: real Jev via OpenRouter (OPENROUTER_API_KEY) or TypeSafe (TYPESAFE_API_KEY), or B: simulation with this agent or an explicitly chosen available model such as DeepSeek. Wait for consent; errors do not authorize switching. Check key presence only, never values. Real calls send evidence and cost money; get approval before sending private data.

For B, skip CLI/API calls. Mark agent_simulation or model_simulation, identify the actual model when available, set jev_called: false, probability: null and confidence: null. Return a value, evidence-based reason and needs_review; use null/review when evidence is missing. Do not invent Jev output or probabilities. Choice uses supplied labels, Noul uses booleans, Score uses integer rubric indices.

For A, use the existing jev-decide CLI with the chosen --provider openrouter or --provider typesafe. If absent, explain the dependency; do not silently install. --dry-run is offline validation, not a judgment. Exit 0 means selected/scored, 2 means review, 1 means error. Read each value: false Noul remains false. Selection is not permission, and confidence is not accuracy. Keep unknown/review paths.

First request

Adapt the example. The shared CLI needs Python 3.10+; no sibling skill is needed. Host tools still own collection and actions. Resolve <skill-dir> to this installed folder:

jev-decide decide <skill-dir>/assets/example.json --dry-run
# After approval, send the edited request with the selected provider:
jev-decide decide /path/to/request.json --provider openrouter

Pick one mode

Browser or desktop: read UI steps; start with the UI request.
Game or simulation: read world steps; start with the world request.

Use only the current mode. A simulated world is not permission to operate a real account. The host validates legal actions, checks freshness and applies the result.

Context and parallelism

Jev does not inherit the agent's history. Include the goal, rules, fresh context, legal candidates and relevant outcomes. Batch independent checks in the same request. Use bounded concurrency only for independent requests; the host owns scheduling. Wait for a new observation after an action before asking a dependent question.

Examples

Next browser action · Browser wait versus intervention · Browser outcome verification

More tasks and local templates. Open only the matching row; there is no need to read the full README before a judgment.

---name: jev-actdescription: Choose one legal next action in a browser, desktop, game or simulation. Supply fresh observed state and available actions. The host or simulator executes and checks the result; selection does not grant permission.---# Choose the next action## Use safelyChoose the service once and keep that choice. If unset, ask **A: real Jev** viaOpenRouter (`OPENROUTER_API_KEY`) or TypeSafe (`TYPESAFE_API_KEY`), or **B: simulation**with this agent or an explicitly chosen available model such as DeepSeek. Wait forconsent; errors do not authorize switching. Check key presence only, never values.Real calls send evidence and cost money; get approval before sending private data.For B, skip CLI/API calls. Mark `agent_simulation` or `model_simulation`, identifythe actual model when available, set `jev_called: false`, `probability: null` and`confidence: null`. Return a value, evidence-based reason and `needs_review`; usenull/review when evidence is missing. Do not invent Jev output or probabilities.Choice uses supplied labels, Noul uses booleans, Score uses integer rubric indices.For A, use the existing `jev-decide` CLI with the chosen `--provider openrouter`or `--provider typesafe`. If absent, explain the dependency; do not silently install.`--dry-run` is offline validation, not a judgment. Exit 0 means selected/scored,2 means review, 1 means error. Read each value: false Noul remains false. Selectionis not permission, and confidence is not accuracy. Keep unknown/review paths.## First requestAdapt [the example](assets/example.json). The shared CLI needs Python 3.10+;no sibling skill is needed. Host tools still own collection and actions.Resolve `<skill-dir>` to this installed folder:```bashjev-decide decide <skill-dir>/assets/example.json --dry-run# After approval, send the edited request with the selected provider:jev-decide decide /path/to/request.json --provider openrouter```## Pick one mode- **Browser or desktop:** read [UI steps](references/ui.md); start with [the UI request](assets/example.json).- **Game or simulation:** read [world steps](references/world.md); start with [the world request](assets/world.json).Use only the current mode. A simulated world is not permission to operate a realaccount. The host validates legal actions, checks freshness and applies the result.## Context and parallelismJev does not inherit the agent's history. Include the goal, rules, fresh context,legal candidates and relevant outcomes. Batch independent checks in the same request.Use bounded concurrency only for independent requests; the host owns scheduling.Wait for a new observation after an action before asking a dependent question.## Examples[Next browser action](https://github.com/wuyoscar/jev-skill#sc-a23) · [Browser wait versus intervention](https://github.com/wuyoscar/jev-skill#sc-a24) · [Browser outcome verification](https://github.com/wuyoscar/jev-skill#sc-a25)[More tasks and local templates](references/scenarios.md). Open only the matching row;there is no need to read the full README before a judgment.

Frequently asked questions

Jev Act