start · wrg32786 — Claude Code skill: setup and use
start — a Claude Code skill that opens a home screen on first launch or when work resumes. It guides setup or gives a brief status update with an action menu.
MIT ★ 26 updated 2026-11-16
npx skills add wrg32786/aigent-os --skill start -a claude-codeWhat start does
The start skill checks whether setup has been completed and chooses the matching flow. On a first run, it records progress in .aigent/state.json, asks questions through /operator-setup, then calls /first-win after an action is chosen and saves the result. Setup becomes ready only after the memory and first artifact have been written.
On return, the skill reads memory/about-you.md and recent session memory, then gives a short briefing and offers actions such as planning the day, drafting content, doing research, or building an automation. The flow also uses /resume, /context-capsule, and /statusline. If those commands are unavailable, their steps in the flow cannot run.
What it helps with
- guide a first-time setup
- summarize current priorities and blockers
- choose a next action from a home screen
- create a first usable result after setup
Example prompts for the agent
- Open the home screen and guide me through initial setup.
- Give me today's briefing and show the action menu.
- Help me pick what to work on next today.
How to install start
1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill start -a claude-code2Manually: put the files in place
~/.claude/skills/start/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/start/
1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill start -a codex2Manually: put the files in place
.agents/skills/start/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill start -a cursor2Manually: put the files in place
.cursor/skills/start/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install wrg32786/aigent-os skills/start2Manually: put the files in place
.github/skills/start/SKILL.md1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill start -a gemini-cli2Platform CLI
gemini skills install https://github.com/wrg32786/aigent-os --path skills/startManually: put the files in place
.devin/skills/start/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill start -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about start
How do I install start in Claude Code?
Run npx skills add wrg32786/aigent-os --skill start -a claude-code. The skill file is at ~/.claude/skills/start/SKILL.md.
What does the start skill do?
It guides first-run setup and starts work on an initial artifact. On later runs, it reads session memory and presents a briefing with an action menu.
Which commands does start use?
The setup and first artifact flow call /operator-setup and /first-win. It also uses /resume, /context-capsule, and /statusline.
Source text
SKILL.md · MIT
---
name: start
description: "Provide a home screen with first-run setup, first win, or a return briefing and action menu. Use when launching for the first time or asking what to do today."
trigger: /start
---
# /start: wake the operator
Use this skill on first launch and whenever the operator asks for the home screen, start menu, or orientation.
## State contract
The canonical first-run state is `.aigent/state.json`:
```json
{
"schemaVersion": 1,
"status": "uninitialized | setup-in-progress | ready",
"completedAt": null
}
```
For compatibility with installations before v0.9, `.aigent/first-run-done` is also recognized.
- `status: ready` or a legacy first-run marker means returning operator.
- Missing state, invalid state, or any other status means first run.
- If the legacy marker exists but the JSON state is not ready, migrate the JSON state to `ready` silently.
## First run
1. Write `status: setup-in-progress` before asking questions, so an interrupted setup can resume.
2. Give a two-line, warm, plain greeting:
> I'm your AIgent operator. Let's get you one real win in the next few minutes.
3. Run `/operator-setup`, one question at a time.
4. When the operator chooses an action from the personal briefing, run `/first-win` and produce a usable artifact.
5. Teach the operating loop once:
> Later sessions pick up on their own via `/resume`. When you are done, say "close up" or run `/context-capsule` so the work is banked (it also autosaves as you go). Run `/statusline` once in Claude Code to show context usage.
6. Only after setup memory and the first artifact are successfully written:
- Set `.aigent/state.json` to `status: ready` with an ISO-8601 `completedAt` value.
- Write `.aigent/first-run-done` for backward compatibility.
Do not mark setup ready before the durable writes succeed.
## Returning operator
1. Read `memory/about-you.md` and recent session memory.
2. Give a short, specific briefing: what matters today, what changed, and what is blocked.
3. Surface the action menu.
## Action menu
Present a short numbered list and accept either the number or a plain-language request:
1. Plan my day
2. Draft content
3. Research something
4. Build an automation
5. Check my numbers
6. Talk it through
On first run, lead with zero-dependency actions so the first win never waits on a connector.
## Rules
- Use plain English and the operator's nouns.
- Ask one setup question at a time.
- Never invent business facts.
- If an operation fails, state what failed, what was attempted, and what durable state was or was not written.
- Do not leave `status: setup-in-progress` indefinitely after a successful setup.
© wrg32786 · MIT · wrg32786/aigent-os
Similar skills
All skills for Agent setup: memory, context, orchestration →start by other authors comes first.