setup · wrg32786 — Claude Code skill: what it does, how to install
This Claude Code skill configures aigent-OS identity, priorities, and authority boundaries. Run /setup to configure the system or revise specific sections later.
MIT ★ 26 updated 2026-11-16
npx skills add wrg32786/aigent-os --skill setup -a claude-codeWhat setup does
The skill guides aigent-OS configuration section by section: role and collaboration style, current priorities, AI authority, decision rules, projects, and people. It can also define specialist agents and store their details for registration in .claude/agents/.
It reads installation state from .aigent/state.json. If the status is not ready, it routes to /start unless advanced setup was explicitly requested; for a ready installation, it asks which section to revise. Completed sections are saved as the interview proceeds, and unrelated user content is preserved. A failed required write prevents the installation from being marked ready.
When configuration is complete, the skill logs the changes, updates state, and creates .aigent/first-run-done. Installing it with the npm command requires npx.
What it helps with
- record the operator’s role and collaboration preferences
- set active priorities and urgent outcomes
- define which AI actions need human approval
- update project and people notes
- write a specialist agent definition
Example prompts for the agent
/setuprevise AI authority boundaries and spending approvals/setupupdate my current priorities and urgent outcome/setupadd a project and define a specialist agent
How to install setup
1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill setup -a claude-code2Manually: put the files in place
~/.claude/skills/setup/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/setup/
1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill setup -a codex2Manually: put the files in place
.agents/skills/setup/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill setup -a cursor2Manually: put the files in place
.cursor/skills/setup/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install wrg32786/aigent-os skills/setup2Manually: put the files in place
.github/skills/setup/SKILL.md1Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill setup -a gemini-cli2Platform CLI
gemini skills install https://github.com/wrg32786/aigent-os --path skills/setupManually: put the files in place
.devin/skills/setup/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add wrg32786/aigent-os --skill setup -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 setup
How do I install setup in Claude Code?
Run npx skills add wrg32786/aigent-os --skill setup -a claude-code. The skill file is available at ~/.claude/skills/setup/SKILL.md.
What does setup do?
It configures aigent-OS identity, priorities, AI authority, decision rules, projects, and people. It can also save definitions for specialist agents.
How is setup different from start?
/start handles the initial onboarding flow. /setup provides deeper configuration and can revise individual sections later.
Source text
SKILL.md · MIT
---
name: setup
description: "Configure aigent-OS identity, priorities, authority, projects, people, and decision logic. Use when running setup, reconfiguring settings, or changing authority levels."
trigger: /setup
---
# aigent-OS setup and reconfiguration
`/start` owns the day-one onboarding arc. `/setup` is the deeper configuration flow and may also be used later to revise one section.
## Entry behavior
Read `.aigent/state.json` first.
- If state is missing or not `ready`, route through `/start` unless the operator explicitly requested advanced setup.
- If state is `ready`, ask which configuration area needs revision rather than replaying the entire interview.
- Never detect setup state by searching for placeholder prose in a Markdown file.
Before writing, set `status: setup-in-progress` while preserving any prior `completedAt`. On interruption, the next `/start` or `/setup` should resume from the last durably completed section.
## Interview
Ask one question at a time and write each completed section immediately.
### 1. Identity
Ask for the operator's role, responsibilities, preferred collaboration style, and risk posture. Update `system/00_identity.md` without replacing framework-owned doctrine.
### 2. Priorities
Ask for the top two or three active priorities, the most urgent outcome, operating mode, and blockers. Update `vault/memory/ACTIVE_PRIORITIES.md`.
### 3. Authority boundaries
Ask what the AI may handle autonomously, what needs confirmation, what is human-only, and whether spending has an escalation threshold. Update `system/12_authority_matrix.md`.
### 4. Decision logic
Ask what makes the operator accept or reject opportunities, their active-work limit, and their most common time trap. Update `system/14_decision_framework.md`.
### 5. Projects and people
For each active project, capture purpose, current state, priority, next action, and relevant people. Create or update notes under `vault/projects/` and `vault/people/`, then connect them from active priorities with wikilinks.
### 6. Specialist agents
Ask whether any specialized agents are useful now. For each accepted agent, capture name, scope, tools, model tier, escalation boundary, and success criteria. Store the definition in `vault/agents/` with valid `name:` and `tools:` frontmatter so the installer can register it in `.claude/agents/`.
## Completion
After all requested sections are durably written:
1. Summarize exactly what changed.
2. Append an idempotent setup entry to `vault/memory/SESSION_LOG.md`.
3. Set `.aigent/state.json` to:
```json
{
"schemaVersion": 1,
"status": "ready",
"completedAt": "<ISO-8601 timestamp>"
}
```
4. Write `.aigent/first-run-done` for compatibility with older installs.
5. Teach or remind the operator of `/open`, normal work, and `/close`.
Do not mark the installation ready when required writes failed.
## Rules
- One section at a time.
- Adapt to information already supplied.
- Mirror the operator's language without inventing details.
- Preserve user-authored content and make precise edits.
- Sections may be deferred explicitly.
- Reconfiguration must not erase unrelated state.
© wrg32786 · MIT · wrg32786/aigent-os
Similar skills
SKILL.md skills catalog →setup by other authors comes first.