# OneShotAiPanel

> Open-source React component that connects to the LLM you configure (opencode, shadcn, fallback) and runs your tasks in a single turn: JSON streaming, schema validation, diff review and direct plug into your stack.

## What is OneShotAiPanel?

A React component that connects to the LLM of your choice and turns a request into validated deliverables, without manual copy-paste.

## Features

- **Connection to your configured LLM** : the component calls the provider you configure (opencode, shadcn, fallback, or a custom onSend)
- **Real-time JSON streaming** : the model response streams in and is validated against schemas (zod)
- **Validated JSON schemas** : each ticket expects a precise, schema-controlled structure
- **Automatic validation** : on mismatch, targeted feedback is sent back for a second turn
- **Diff & review** : when content already exists, a line-by-line diff shows what changes before integrating
- **Direct plug** : retrieve the selected content via onPlug and wire it into your code, adapter or SDK

## How it works

- **1. Describe the task** — A user prompt, a system prompt and context files. The component assembles the final prompt.
- **2. Configure the model connection** — The component connects to the provider you configure via the adapter (opencode, shadcn, fallback) or your own onSend. Without this connection, nothing runs.
- **3. A single send** — The model responds with streaming JSON. Each ticket expects a precise, schema-controlled structure.
- **4. Automatic validation** — The response is validated against the schemas. On mismatch, targeted feedback is sent back for a second turn.
- **5. Diff & review** — If content already exists, a line-by-line diff shows you what changes before integrating.
- **6. Plug** — Retrieve the selected content via onPlug and wire it into your code, adapter or SDK.

## Configuration (main props)

### Content

- `title` — Title shown in the component header.
- `systemPrompt` — The role and rules given to the model.
- `initialUserPrompt` — The initial request pre-filled in the user field.

### Context

- `files` — Project files provided as context to the model.
- `tickets` — Task breakdown, each ticket with its JSON schema.

### Contract

- `parser` — Customize response parsing and validation.
- `invalidMode` — Warn or block when the response does not validate the schemas.

### Adapter

- `adapter` — Provider (opencode, shadcn, fallback), model and API key.
- `onSend` — Replace the send with your own logic (SDK, proxy…).

### Events

- `onPlug` — Called when the user integrates the selected contents.

### i18n

- `language` — Interface language.
- `labels` — Fine-grained override of any label.

### UI

- `actionLabel` — Label of the main run button.
- `showInfoButton` — Shows the help / integration / credits button.
- `showSettingsButton` — Shows the settings button.
- `showInfoIntegration` — Shows the integration point in the help panel.
- `showInfoCredits` — Shows the credits in the help panel.

## Installation

The source code and demo live on GitHub:

- Repository : https://github.com/bn-dev-19/one-shot-ai-panel
- Current version : v1.1.3 (archive : https://github.com/bn-dev-19/one-shot-ai-panel/archive/refs/tags/v1.1.3.zip)

## Playground

The component actually runs on the site: : https://oneshot-ai-panel.asistem19.com/en#playground

## Links

- Home : https://oneshot-ai-panel.asistem19.com/en
- GitHub : https://github.com/bn-dev-19/one-shot-ai-panel
- License : https://oneshot-ai-panel.asistem19.com/en/license
- Terms of use : https://oneshot-ai-panel.asistem19.com/en/terms
- Privacy policy : https://oneshot-ai-panel.asistem19.com/en/privacy
- Credits : https://oneshot-ai-panel.asistem19.com/en/credits