Jarvis

The sidekick thread Mike talks to about all his other threads. It hands out work, hears back when results come in, and translates results he cannot eyeball. Use when the user says jarvis, or asks where to start, to fire agents and hear back, or to oversee one lane.

10 stars 1 forksWriting & Content

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add thatmike1/claude-skills --skill jarvis
Manual install
git clone --depth 1 https://github.com/thatmike1/claude-skills.git /tmp/claude-skills
mkdir -p ~/.claude/skills
cp -r /tmp/claude-skills/jarvis ~/.claude/skills/jarvis

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# jarvis

You are the thread Mike talks to *about* his other threads. He fires work as ideas
arrive, comes back later and is lost; you make coming back cheap. The shared rules
already cover how any thread delegates, briefs and reports; this skill is only what
the role adds on top.

**Voice.** At most one dry, wry line per message, never inside an item that needs
him. The Jarvis feel comes from timing and restraint, not phrasing.

**Your context is the day's memory.** Follow the shared delegation rule, quick-job
exceptions included; you hold every lane at once, so longer work goes to an agent even
mid-conversation. Give each task its own `clientRequestId` and end its prompt with
the paragraph from `node ~/.claude/skills/jarvis/scripts/jarvis.mjs prompt-tail`.

## Digests

When something needs him or everything sent out is back, write one full digest that
replaces the earlier ones, in this order:

1. The count: "Three back. One needs you."
2. *Waiting on you*: every action open to him right now, earlier ones included, each
   restated in full in a short paragraph: what it is, what he decides.
3. Done, one line each.
4. Running. Work still failing checks is running, not waiting on him.
5. *Later*: anything deliberately deferred to another day.

## GPT reviews

A GPT review gets the prompt in `references/adversarial-review.md` pasted in with
target and focus filled. Its final message is a JSON verdict with findings (file,
lines, confidence, fix); parse it before relaying it.
Read the full skill on GitHub

Details

Skill name
jarvis
Path
jarvis/SKILL.md
Repo updated
18 November 2026

FAQ

What does the Jarvis skill do?

The sidekick thread Mike talks to about all his other threads. It hands out work, hears back when results come in, and translates results he cannot eyeball. Use when the user says jarvis, or asks where to start, to fire agents and hear back, or to oversee one lane.

How do I install the Jarvis skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add thatmike1/claude-skills --skill jarvis. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Jarvis popular?

It lives in thatmike1/claude-skills, which has 10 GitHub stars and 1 forks.

Related skills

Afk Prompt

10

thatmike1/claude-skills

Write a copy-paste prompt that hands one or more tasks (usually beads issues) to a fresh Claude instance to run autonomously while the user is away — no questions, no waiting for input. Includes picking which tasks are safe to run unattended. Use when the user says they're going AFK (shower, lunch, errand, overnight), asks "which tasks can just run without me", or wants an autonomous/batch prompt. For attended sessions the user will steer, use the live-prompt skill instead.

AI, LLMs & Agents

AI CV Scanner

10

thatmike1/claude-skills

Mine Claude Code conversation history to answer questions about your AI experience. Scans session indexes, spawns parallel subagents to deep-dive relevant conversations, and drafts evidence-based answers. Use when user mentions CV, AI experience, questionnaire, scan my conversations, or needs to document their AI usage.

AI, LLMs & Agents

Arena

10

thatmike1/claude-skills

Give several T3 child candidates the same prompt, rank them against a rubric written before they run, pick a base and graft the best of the rest into it. Run with /arena when one attempt at a non-trivial artifact would lock in the wrong shape.

AI, LLMs & Agents

Md2wechat

9

lingengyuan/my-skills

Convert Markdown articles into WeChat Official Account compatible inline-styled HTML and optionally publish them to the WeChat draft box. Use when formatting articles for WeChat, rendering WeChat HTML, styling Markdown with themes, uploading article images, or publishing drafts to WeChat.

Writing & Content

Spec to Sequence

9

pasqal-io/agent-marketplace

Generate a Pulser sequence builder for a neutral-atom (Rydberg) experiment from an experiment_spec.json. Output is a *_sequence.py with two standardised functions — build_sequence() and compute_observable() — consumed by validate-emu, qpu-submit and harvest-and-analyze. Triggered by phrases like "generate the Pulser sequence", "write the sequence file for this spec", "spec to sequence", "turn the spec into Pulser code".

Writing & Content

Publication Ready Docs

9

lingengyuan/my-skills

撰写和修订正式工程技术文档(技术方案、详细设计、实施计划、README、API 文档、运行手册)。先查证源码与既有设计,删除冗余和过程性内容,保留必要的规则、改动对象、关键代码和配置。

Writing & Content

Browse the most popular Claude skills, all writing & content skills, or write your own with the skills generator.