agentsclimarketplace

Opt chat install

Skill reopt-ai/reopt-skills/skills/opt-chat-install

Install or upgrade @reopt-ai/opt-chat in a consumer project. Auto-branches by current install state. Triggers on "opt-chat install", "opt-chat init", "opt-chat setup", "chat install", "install chat", "set up AI chat", "opt-chat upgrade", "opt-chat update", "chat update".From its SKILL.md

Install
npx -y skills add reopt-ai/reopt-skills --skill opt-chat-install

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
  • runs commandsInstructs the agent to run 2 commands, including `npm config get @reopt-ai:registry` and 1 more.

SKILL.md

4.6 KB, ~1.2k tokens by cl100k_base, as published. Nobody here has run it

opt-chat Install

This is NOT the opt-chat you know. opt-chat ships no dist/docs/ — read node_modules/@reopt-ai/opt-chat/README.md before writing code.

When to apply

Consumer project depends on @reopt-ai/opt-chat. Triggers: "install", "init", "setup", "upgrade", "update" — with chat / opt-chat / "set up AI chat".

What opt-chat provides

AreaPattern
CoreConversation, Message, PromptInput (composable)
HookuseChatSession (Vercel AI SDK wrapper)
Parts28+ message part renderers (reasoning, tool, artifact, code, agent, …)
InputAttachments, SpeechInput, ModelSelector, AudioPlayer / VoiceSelector
Streamingstreamdown + shiki syntax highlighting
Flow@reopt-ai/opt-chat/flow agent-graph Canvas (optional peer @xyflow/react)
Stylingopt-ui global CSS or @reopt-ai/opt-chat/styles.css (CSS-variable tokens) — no Tailwind config required

Invocation

/opt-chat-install              # Auto-branch (missing → init, installed → upgrade)
/opt-chat-install --upgrade    # Explicit upgrade
/opt-chat-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-chat does not, as of 1.0.0). Fallback: agent-rules.md bundled with this skill. Wrap content between:

<!-- BEGIN:reopt/opt-chat-agent-rules -->
…content from source…
<!-- END:reopt/opt-chat-agent-rules -->

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. The “.npmrc 등록 후” comment in the bundled 1.0.0 README is stale setup text; do not follow it.

  2. Prereqs — Node 18+, React 19+, a Vercel AI SDK–compatible AI endpoint. No Tailwind required. Optional peers: react-jsx-parser (JSXPreview parts), @xyflow/react (/flow Canvas).

  3. App wiring — properties of the consumer app:

    • Styles: import "@reopt-ai/opt-chat/styles.css" at the app root unless opt-ui's global CSS is already loaded (they share keyframes / data-attribute tokens).
    • AI endpoint route (e.g. app/api/chat/route.ts) returning a Vercel AI SDK stream.
    • Default Chat component composition (Conversation > Message[] > PromptInput).

Step 3 — Route to module docs

opt-chat ships no dist/docs/. Route to node_modules/@reopt-ai/opt-chat/README.md sections (and CHANGELOG.md).

Task signalRead
Quick start + composition (Conversation / Message / PromptInput)README.md §§ Quick start, Component catalog
Hooks (useChatSession), 28+ part renderers, input (Attachments / SpeechInput / ModelSelector), streamingREADME.md § Component catalog
Styling / CSS import (opt-ui CSS or styles.css)README.md § Styles
Flow / agent-graph Canvas (@reopt-ai/opt-chat/flow) + optional peersREADME.md §§ Optional peers, Flow
opt-ui token / design-system integrationREADME.md § Design system integration
Version migration / breaking changesREADME.md § Migration, CHANGELOG.md

Pipeline (auto-branch)

#StepInitUpgrade
1Detect current state
2Public-registry preflight + legacy override cleanup
3Install / update package
4Styles import check (opt-ui CSS or styles.css)
5AI SDK endpoint check
6Generate default Chat component
7Breaking-change edits
8Verify + summary

Safety

  • Never upgrade without an impact scan.
  • Apply breaking-change edits in logical groups.
  • Do not finish until tsc --noEmit passes.
  • Never commit — do not commit or push without an explicit request from the user.

Verify

  1. npx tsc --noEmit passes.
  2. Chat conversation renders, AI endpoint streams back, parts render correctly (text + at least one code block).

What ships with it: 3 files

2.2 KB alongside SKILL.md

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.