# Opt Shell Install

> |

- **Type:** Skill
- **Install:** `agentstack add skill-reopt-ai-reopt-skills-opt-shell-install`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [reopt-ai](https://agentstack.voostack.com/s/reopt-ai)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [reopt-ai](https://github.com/reopt-ai)
- **Source:** https://github.com/reopt-ai/reopt-skills/tree/main/skills/opt-shell-install
- **Website:** https://skills.sh/reopt-ai/reopt-skills

## Install

```sh
agentstack add skill-reopt-ai-reopt-skills-opt-shell-install
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# opt-shell Install

> This is NOT the opt-shell you know. `@reopt-ai/opt-shell` (formerly `@reopt-ai/opt-harness`) is the runtime product-frame layer between opt-ui and product screens. It ships **no** `dist/docs/` — read `node_modules/@reopt-ai/opt-shell/shell-llms.txt` (agent guide) and `README.md` before writing code.

## When to apply

Consumer project depends on `@reopt-ai/opt-shell`. Triggers: "install", "init", "setup", "upgrade", "update" — with `shell` / `opt-shell` (or the legacy `harness` / `opt-harness`). `@reopt-ai/opt-harness` was renamed to `@reopt-ai/opt-shell`; if a project still references `opt-harness`, migrate the dependency first.

## What opt-shell provides

| Area | Pattern |
|---|---|
| Workspace recipes | `DashboardWorkspace`, `ListWorkspace`, `DetailWorkspace`, `EditorWorkspace`, landing — pick via the decision tree in `shell-llms.txt` |
| Non-recipe surface | `ShellFullscreenToolSurface` (full-viewport tools — code editor, canvas) |
| Policy | density, contentWidth (`full` / `wide` / `normal` / `narrow`), navigationMode, motion |
| Adapters | data-engine wrappers supplying loading / empty / error chrome |
| State UX | shared state boundaries; every recipe requires `header` + `content` slots |
| Authoring audit | `@reopt-ai/opt-cli/audit` + `opt harness` commands; published opt-shell 1.0.0 has no `./audit` export |

## Invocation

```
/opt-shell-install              # Auto-branch (missing → init, installed → upgrade)
/opt-shell-install --upgrade    # Explicit upgrade
/opt-shell-install --check      # Analyze only
```

## Step 1 — Pin agent rules into AGENTS.md / CLAUDE.md

Source: the module's own agent-rules file once it ships one (`@reopt-ai/opt-shell` ships `shell-llms.txt`, an agent guide, but not a marker-block file as of 1.0.0). Fallback: `agent-rules.md` bundled with this skill. Wrap content between:

```

…content from source…

```

**Idempotent:** replace only between markers.

## Step 2 — Consumer-side setup (this skill owns; docs cannot)

1. **Public npm registry** — no token or scoped `.npmrc` entry is required. Inspect the project `.npmrc` and `npm config get @reopt-ai:registry`; if the scope still resolves to GitHub Packages, remove only the legacy project entry `@reopt-ai:registry=https://npm.pkg.github.com`. Preserve unrelated registry/auth settings, and ask before changing user/global npm config.

2. **Peers** — opt-shell's `peerDependencies` (install / run their skills first if missing):
   - **Required:** `@reopt-ai/opt-palette` (theme engine), `react` / `react-dom` 19+
   - **Optional** (only if you use that adapter): `@reopt-ai/opt-datagrid` (`/opt-datagrid-install`), `@reopt-ai/opt-editor` (`/opt-editor-install`), `@reopt-ai/opt-calendar`

   opt-ui tokens come in transitively through opt-palette — run `/opt-ui-install` if the theme layer is missing.

3. **App wiring** — properties of the consumer app:
   - A workspace recipe at the screen root (`header` + `content` slots are mandatory).
   - A shell manifest / policy config (density, contentWidth, navigationMode, motion).
   - Adapters wired to your data engines wherever loading / empty / error chrome is needed.

## Step 3 — Route to module docs

opt-shell ships **no** `dist/docs/`. Route to `shell-llms.txt` (agent guide) and `README.md`.

| Task signal | Read |
|---|---|
| Recipe decision tree, slot rules, policy | `shell-llms.txt` |
| Component / prop API, exports (`.`, `./core`, `./meta`) | `README.md` |
| Authoring audit / scoring | `@reopt-ai/opt-cli/audit`; `npx @reopt-ai/opt-cli harness check|test|doctor` |
| Breaking changes per version | `CHANGELOG.md` |

## Pipeline (auto-branch)

| # | Step | Init | Upgrade |
|---|---|---|---|
| 1 | Detect current state (incl. legacy `opt-harness` dep) | ✓ | ✓ |
| 2 | Public-registry preflight + legacy override cleanup | ✓ | ✓ |
| 3 | Install / update package | ✓ | ✓ |
| 4 | Peer check (opt-palette required; opt-datagrid / opt-editor / opt-calendar optional) | ✓ | ✓ |
| 5 | Shell manifest / policy config | ✓ | – |
| 6 | First workspace recipe | ✓ | – |
| 7 | Breaking-change edits | – | ✓ |
| 8 | Verify + summary | ✓ | ✓ |

## Safety

- Never upgrade without an impact scan.
- Confirm before overwriting existing files.
- Apply breaking-change edits in logical groups, never bulk.
- Do not finish until `tsc --noEmit` passes.
- **Never commit** — do not commit or push without an explicit request.

## Verify

1. `npx tsc --noEmit` passes.
2. App boots into the chosen workspace recipe (`header` + `content` slots filled); policy (density / contentWidth / navigationMode) applies.
3. (If an adapter is used) the data-grid or editor adapter renders the peer's content with shell chrome (loading / empty / error states).

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [reopt-ai](https://github.com/reopt-ai)
- **Source:** [reopt-ai/reopt-skills](https://github.com/reopt-ai/reopt-skills)
- **License:** MIT
- **Homepage:** https://skills.sh/reopt-ai/reopt-skills

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-reopt-ai-reopt-skills-opt-shell-install
- Seller: https://agentstack.voostack.com/s/reopt-ai
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.