### Clone Repository and Run Example

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/ai-sdk.mdx

Steps to clone the Agentic repository, install dependencies, build the project, set up environment variables, and run the AI SDK example.

```sh
git clone git@github.com:transitive-bullshit/agentic.git
cd agentic
pnpm install
pnpm build
echo 'OPENAI_API_KEY=your-key' >> .env
npx tsx examples/ts-sdks/ai-sdk/bin/weather.ts
```

--------------------------------

### Install @agentic/ai-sdk

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/stdlib/ai-sdk/readme.md

Install the @agentic/ai-sdk package along with zod and ai using npm. This is the initial setup step for using the SDK.

```bash
npm i @agentic/ai-sdk zod ai
```

--------------------------------

### Clone Repository and Run Example

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/mastra.mdx

Steps to clone the Agentic repository, install dependencies, build the project, set up environment variables, and run the Mastra weather agent example.

```sh
git clone git@github.com:transitive-bullshit/agentic.git
cd agentic
pnpm install
pnpm build
echo 'OPENAI_API_KEY=your-key' >> .env
npx tsx examples/ts-sdks/mastra/bin/weather.ts
```

--------------------------------

### Run the OpenAI Example

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/openai-responses.mdx

Clone the agentic repository, install dependencies, build the project, set your OpenAI API key in a .env file, and then run the example using tsx.

```sh
git clone git@github.com:transitive-bullshit/agentic.git
cd agentic
pnpm install
pnpm build
echo 'OPENAI_API_KEY=your-key' >> .env
npx tsx examples/ts-sdks/openai/bin/weather-responses.ts
```

--------------------------------

### Run the example project

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/openai-chat.mdx

Commands to clone the repository and execute the weather example script.

```sh
git clone git@github.com:transitive-bullshit/agentic.git
cd agentic
pnpm install
pnpm build
echo 'OPENAI_API_KEY=your-key' >> .env
npx tsx examples/ts-sdks/openai/bin/weather.ts
```

--------------------------------

### Run the LlamaIndex example

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/llamaindex.mdx

Clone the repository and execute the weather example script.

```sh
git clone git@github.com:transitive-bullshit/agentic.git
cd agentic
pnpm install
pnpm build
echo 'OPENAI_API_KEY=your-key' >> .env
npx tsx examples/ts-sdks/llamaindex/bin/weather.ts
```

--------------------------------

### Clone Agentic Repository and Run LangChain Example

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/langchain.mdx

Steps to clone the Agentic repository, install dependencies, build the project, set up environment variables, and run the LangChain example using tsx.

```sh
git clone git@github.com:transitive-bullshit/agentic.git
cd agentic
pnpm install
pnpm build
echo 'OPENAI_API_KEY=your-key' >> .env
npx tsx examples/ts-sdks/langchain/bin/weather.ts
```

--------------------------------

### Install @agentic/serpapi and zod

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/stdlib/serpapi/readme.md

Use npm to install the @agentic/serpapi package along with zod for schema validation. This is the primary setup step for using the SDK.

```bash
npm i @agentic/serpapi zod
```

--------------------------------

### Start All Services in Development Mode

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/CLAUDE.md

Use this command to start all services concurrently for local development.

```bash
pnpm dev
```

--------------------------------

### Install @agentic/platform-tool-client

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/packages/tool-client/readme.md

Install the package using npm.

```bash
npm i @agentic/platform-tool-client
```

--------------------------------

### Install @agentic/platform-api-client

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/packages/api-client/readme.md

Install the package using npm. This is the first step before using the client in your project.

```bash
npm i @agentic/platform-api-client
```

--------------------------------

### Install Agentic CLI with bun

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/existing-mcp-server.mdx

Install the Agentic CLI globally using bun. Ensure Node.js is installed.

```bash
bun install -g @agentic/cli
```

--------------------------------

### Install Agentic Platform with bun

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/existing-mcp-server.mdx

Install the `@agentic/platform` package as a development dependency using bun.

```bash
bun add -d @agentic/platform
```

--------------------------------

### Importing External ESM Module Example

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/CLAUDE.md

Illustrates importing an external module that uses ESM, like Zod. Ensure the module is installed.

```javascript
import zod from 'zod'
```

--------------------------------

### Install @agentic/langchain dependencies

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/stdlib/langchain/readme.md

Run this command to install the necessary packages for using the adapter.

```bash
npm i @agentic/langchain zod @langchain/core
```

--------------------------------

### Install additional dependencies

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/genkit.mdx

Install optional packages for OpenAI support.

```npm
npm install genkitx-openai dotenv
```

```pnpm
pnpm add genkitx-openai dotenv
```

```bun
bun add genkitx-openai dotenv
```

```yarn
yarn add genkitx-openai dotenv
```

--------------------------------

### Install LangChain and Agentic Packages (bun)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/langchain.mdx

Install the necessary LangChain and Agentic packages using bun.

```bash
bun add langchain @langchain/core @langchain/agents @agentic/langchain @agentic/platform-tool-client
```

--------------------------------

### Install Additional Dependencies (bun)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/mastra.mdx

Install the OpenAI provider and dotenv for environment variable management using bun.

```bash
bun add @ai-sdk/openai dotenv
```

--------------------------------

### Install Mastra Core Packages (bun)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/mastra.mdx

Install the necessary Mastra core packages using bun.

```bash
bun add @mastra/core @agentic/mastra @agentic/platform-tool-client
```

--------------------------------

### Install fastmcp package

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/py-fastmcp.mdx

Install the fastmcp package using either uv or pip. Ensure Python is installed beforehand.

```bash
uv add fastmcp
```

```bash
pip install fastmcp
```

--------------------------------

### Install @agentic/cli

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/packages/cli/readme.md

Global installation command for the CLI tool.

```bash
npm i -g @agentic/cli
```

--------------------------------

### Create and Execute a LangChain Agent with Tool Calling

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/langchain.mdx

This TypeScript example demonstrates how to create a LangChain agent that uses Agentic tools. It requires an OpenAI API key and an Agentic API key for full functionality. Ensure you have the necessary packages installed.

```typescript
import 'dotenv/config'

import { createLangChainTools } from '@agentic/langchain'
import { AgenticToolClient } from '@agentic/platform-tool-client'
import { ChatPromptTemplate } from '@langchain/core/prompts'
import { ChatOpenAI } from '@langchain/openai'
import { AgentExecutor, createToolCallingAgent } from 'langchain/agents'

async function main() {
  const searchTool = await AgenticToolClient.fromIdentifier('@agentic/search')

  const tools = createLangChainTools(searchTool)
  const agent = createToolCallingAgent({
    llm: new ChatOpenAI({ model: 'gpt-4o-mini', temperature: 0 }),
    tools,
    prompt: ChatPromptTemplate.fromMessages([
      ['system', 'You are a helpful assistant. Be as concise as possible.'],
      ['placeholder', '{chat_history}'],
      ['human', '{input}'],
      ['placeholder', '{agent_scratchpad}']
    ])
  })

  const agentExecutor = new AgentExecutor({
    agent,
    tools
    // verbose: true
  })

  const result = await agentExecutor.invoke({
    input: 'What is the weather in San Francisco?'
  })

  console.log(result.output)
}

await main()
```

--------------------------------

### Install Agentic CLI with pnpm

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/existing-mcp-server.mdx

Install the Agentic CLI globally using pnpm. Ensure Node.js is installed.

```bash
pnpm i -g @agentic/cli
```

--------------------------------

### Install Additional Dependencies (npm)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/mastra.mdx

Install the OpenAI provider and dotenv for environment variable management using npm.

```bash
npm install @ai-sdk/openai dotenv
```

--------------------------------

### Install Agentic CLI with yarn

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/existing-mcp-server.mdx

Install the Agentic CLI globally using yarn. Ensure Node.js is installed.

```bash
yarn global add @agentic/cli
```

--------------------------------

### Install Agentic Platform with pnpm

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/existing-mcp-server.mdx

Install the `@agentic/platform` package as a development dependency using pnpm.

```bash
pnpm add -D @agentic/platform
```

--------------------------------

### Install @agentic/platform-core

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/packages/platform-core/readme.md

Install the package via npm.

```bash
npm i @agentic/platform-core
```

--------------------------------

### Importing from Shared Packages Example

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/CLAUDE.md

Demonstrates importing modules from shared packages using a path alias. Ensure the alias is correctly configured.

```javascript
import { logger } from '~/types'
```

--------------------------------

### Install Additional Dependencies for OpenAI Support (bun)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/langchain.mdx

Install the @langchain/openai and dotenv packages using bun.

```bash
bun add @langchain/openai dotenv
```

--------------------------------

### Install @agentic/platform-types

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/packages/types/readme.md

Install the package using npm.

```bash
npm i @agentic/platform-types
```

--------------------------------

### Create xmcp App

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/ts-xmcp.mdx

Use this command to initiate a new xmcp application. Ensure Node.js is installed prior to execution.

```bash
npx create-xmcp-app@latest
```

--------------------------------

### Install LangChain and Agentic Packages (yarn)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/langchain.mdx

Install the necessary LangChain and Agentic packages using yarn.

```bash
yarn add langchain @langchain/core @langchain/agents @agentic/langchain @agentic/platform-tool-client
```

--------------------------------

### Install LangChain and Agentic Packages (npm)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/langchain.mdx

Install the necessary LangChain and Agentic packages using npm.

```bash
npm install langchain @langchain/core @langchain/agents @agentic/langchain @agentic/platform-tool-client
```

--------------------------------

### Install Additional Dependencies (pnpm)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/mastra.mdx

Install the OpenAI provider and dotenv for environment variable management using pnpm.

```bash
pnpm add @ai-sdk/openai dotenv
```

--------------------------------

### Install Agentic and OpenAI dependencies

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/openai-chat.mdx

Install the necessary packages using your preferred package manager.

```npm
npm install openai @agentic/platform-tool-client
```

```pnpm
pnpm add openai @agentic/platform-tool-client
```

```bun
bun add openai @agentic/platform-tool-client
```

```yarn
yarn add openai @agentic/platform-tool-client
```

--------------------------------

### Install Mastra Core Packages (yarn)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/mastra.mdx

Install the necessary Mastra core packages using yarn.

```bash
yarn add @mastra/core @agentic/mastra @agentic/platform-tool-client
```

--------------------------------

### Install Additional Dependencies (yarn)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/mastra.mdx

Install the OpenAI provider and dotenv for environment variable management using yarn.

```bash
yarn add @ai-sdk/openai dotenv
```

--------------------------------

### Install OpenAI provider dependencies

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/llamaindex.mdx

Install the optional OpenAI provider and dotenv for environment variable management.

```npm
npm install @llamaindex/openai dotenv
```

```pnpm
pnpm add @llamaindex/openai dotenv
```

```bun
bun add @llamaindex/openai dotenv
```

```yarn
yarn add @llamaindex/openai dotenv
```

--------------------------------

### Install Agentic Platform with npm

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/existing-mcp-server.mdx

Install the `@agentic/platform` package as a development dependency using npm.

```bash
npm i -save-dev @agentic/platform
```

--------------------------------

### Install LangChain and Agentic Packages (pnpm)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/langchain.mdx

Install the necessary LangChain and Agentic packages using pnpm.

```bash
pnpm add langchain @langchain/core @langchain/agents @agentic/langchain @agentic/platform-tool-client
```

--------------------------------

### Install @agentic/serper

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/stdlib/serper/readme.md

Install the package along with the required zod peer dependency.

```bash
npm i @agentic/serper zod
```

--------------------------------

### Install Agentic AI SDK and Platform Tool Client

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/ai-sdk.mdx

Install the necessary packages for using Agentic tools with the Vercel AI SDK using npm, pnpm, bun, or yarn.

```bash
npm install ai @agentic/ai-sdk @agentic/platform-tool-client
```

```bash
pnpm add ai @agentic/ai-sdk @agentic/platform-tool-client
```

```bash
bun add ai @agentic/ai-sdk @agentic/platform-tool-client
```

```bash
yarn add ai @agentic/ai-sdk @agentic/platform-tool-client
```

--------------------------------

### ESM Import Example

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/CLAUDE.md

Demonstrates correct ESM import syntax without file extensions. Use for importing modules within the project.

```javascript
import { Foo } from './foo'
```

--------------------------------

### Install Additional Dependencies for OpenAI Support (yarn)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/langchain.mdx

Install the @langchain/openai and dotenv packages using yarn.

```bash
yarn add @langchain/openai dotenv
```

--------------------------------

### Install Mastra Core Packages (npm)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/mastra.mdx

Install the necessary Mastra core packages using npm.

```bash
npm install @mastra/core @agentic/mastra @agentic/platform-tool-client
```

--------------------------------

### Install @agentic/genkit and zod

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/stdlib/genkit/readme.md

Run this command in your terminal to install the necessary packages for using the Genkit adapter.

```bash
npm i @agentic/genkit zod
```

--------------------------------

### Install Additional Dependencies for OpenAI Support (pnpm)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/langchain.mdx

Install the @langchain/openai and dotenv packages using pnpm.

```bash
pnpm add @langchain/openai dotenv
```

--------------------------------

### Install @agentic/llamaindex

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/stdlib/llamaindex/readme.md

Install the @agentic/llamaindex package along with its dependencies zod and llamaindex using npm.

```bash
npm i @agentic/llamaindex zod llamaindex
```

--------------------------------

### Install Mastra Core Packages (pnpm)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/mastra.mdx

Install the necessary Mastra core packages using pnpm.

```bash
pnpm add @mastra/core @agentic/mastra @agentic/platform-tool-client
```

--------------------------------

### Install @agentic/mastra

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/stdlib/mastra/readme.md

Install the @agentic/mastra package along with its dependencies zod and @mastra/core using npm.

```bash
npm i @agentic/mastra zod @mastra/core
```

--------------------------------

### Free Monthly Pricing Plan (TypeScript)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/config/examples.mdx

Example configuration for a free monthly pricing plan, used by default if no plans are specified. Includes rate limiting.

```typescript
import { defineConfig } from '@agentic/platform'

export default defineConfig({
  name: 'Free Monthly Pricing Example',
  description:
    "This example shows the free monthly pricing plan which is used by default for projects that don't specify any pricing plans.",
  origin: {
    type: 'mcp',
    url: 'https://agentic-basic-mcp-test.onrender.com/mcp'
  },
  pricingPlans: [
    {
      name: 'Free',
      slug: 'free',
      lineItems: [
        {
          slug: 'base',
          usageType: 'licensed',
          amount: 0
        }
      ],
      rateLimit: {
        enabled: true,
        interval: 60,
        limit: 1000
      }
    }
  ]
})
```

--------------------------------

### Install Additional Dependencies for OpenAI Provider

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/ai-sdk.mdx

Install the OpenAI provider and dotenv for managing environment variables. This is optional if you are not using OpenAI.

```bash
npm install @ai-sdk/openai dotenv
```

```bash
pnpm add @ai-sdk/openai dotenv
```

```bash
bun add @ai-sdk/openai dotenv
```

```bash
yarn add @ai-sdk/openai dotenv
```

--------------------------------

### ESM Import with Types Example

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/CLAUDE.md

Shows how to import types using ESM syntax without file extensions. Useful for type definitions.

```javascript
import { type Route } from './types/root'
```

--------------------------------

### Install LlamaIndex Agentic dependencies

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/llamaindex.mdx

Install the core LlamaIndex and Agentic packages required for integration.

```npm
npm install llamaindex @llamaindex/workflow @agentic/llamaindex @agentic/platform-tool-client
```

```pnpm
pnpm add llamaindex @llamaindex/workflow @agentic/llamaindex @agentic/platform-tool-client
```

```bun
bun add llamaindex @llamaindex/workflow @agentic/llamaindex @agentic/platform-tool-client
```

```yarn
yarn add llamaindex @llamaindex/workflow @agentic/llamaindex @agentic/platform-tool-client
```

--------------------------------

### Initialize and Use AgenticApiClient

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/packages/api-client/readme.md

Import and instantiate the AgenticApiClient. An optional API key can be provided during initialization. This example demonstrates fetching a public project by its identifier.

```typescript
import { AgenticApiClient } from '@agentic/platform-api-client'

// Takes an optional `apiKey`
const agenticApiClient = new AgenticApiClient()

const searchProject = await agenticApiClient.getPublicProjectByIdentifier({
  projectIdentifier: '@agentic/search'
})
console.log(searchProject)
```

--------------------------------

### Install FastMCP dependencies

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/ts-fastmcp.mdx

Install the required fastmcp and zod packages using your preferred package manager.

```bash
npm add fastmcp zod
```

```bash
pnpm add fastmcp zod
```

```bash
bun add fastmcp zod
```

```bash
yarn add fastmcp zod
```

--------------------------------

### Install Hono Dependencies

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/ts-openapi-hono.mdx

Installs necessary Hono, Zod, and OpenAPI related packages using npm, pnpm, bun, or yarn.

```bash
npm add hono @hono/node-server @hono/zod-openapi zod
```

```bash
pnpm add hono @hono/node-server @hono/zod-openapi zod
```

```bash
bun add hono @hono/node-server @hono/zod-openapi zod
```

```bash
yarn add hono @hono/node-server @hono/zod-openapi zod
```

--------------------------------

### SerperClient TypeScript Example

Source: https://context7.com/transitive-bullshit/chatgpt-api/llms.txt

Example of using the SerperClient directly as a TypeScript client for Google Search, including searching for news, images, videos, and places. Requires SERPER_API_KEY environment variable.

```typescript
import { SerperClient } from '@agentic/serper'
import { createAISDKTools } from '@agentic/ai-sdk'
import { generateText } from 'ai'
import { openai } from '@ai-sdk/openai'

// Create a Serper client (requires SERPER_API_KEY env var)
const serper = new SerperClient({
  apiKey: process.env.SERPER_API_KEY
})

// Use directly as a TypeScript client
const searchResults = await serper.search({
  q: 'latest AI developments',
  num: 10,
  type: 'news'
})
console.log(searchResults.news)

// Search for images
const images = await serper.searchImages('cute puppies')
console.log(images.images)

// Search for videos
const videos = await serper.searchVideos('TypeScript tutorial')
console.log(videos.videos)

// Search for places
const places = await serper.searchPlaces('coffee shops near me')
console.log(places.places)

// Use as an LLM tool
const result = await generateText({
  model: openai('gpt-4o-mini'),
  tools: createAISDKTools(serper),
  toolChoice: 'required',
  prompt: 'Search for the latest news about AI'
})

console.log(result.toolResults[0])
```

--------------------------------

### Install Additional Dependencies for OpenAI Support (npm)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/langchain.mdx

Install the @langchain/openai and dotenv packages using npm to enable OpenAI integration.

```bash
npm install @langchain/openai dotenv
```

--------------------------------

### Use Agentic tools with OpenAI Chat Completions

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/openai-chat.mdx

Example showing how to initialize an Agentic tool, invoke it via OpenAI's tool calling API, and process the result.

```ts
import 'dotenv/config'

import { AgenticToolClient } from '@agentic/platform-tool-client'
import OpenAI from 'openai'

async function main() {
  const searchTool = await AgenticToolClient.fromIdentifier('@agentic/search')
  const openai = new OpenAI()

  const messages: OpenAI.ChatCompletionMessageParam[] = [
    {
      role: 'system',
      content: 'You are a helpful assistant. Be as concise as possible.'
    },
    { role: 'user', content: 'What is the weather in San Francisco?' }
  ]

  {
    // First call to OpenAI to invoke the tool
    const res = await openai.chat.completions.create({
      messages,
      model: 'gpt-4o-mini',
      temperature: 0,
      tools: searchTool.functions.toolSpecs,
      tool_choice: 'required'
    })

    const message = res.choices[0]!.message!
    const toolCall = message.tool_calls![0]!.function!
    const toolResult = await searchTool.callTool(
      toolCall.name,
      toolCall.arguments
    )

    messages.push(message)
    messages.push({
      role: 'tool',
      tool_call_id: message.tool_calls![0]!.id,
      content: JSON.stringify(toolResult)
    })
  }

  {
    // Second call to OpenAI to generate a text response
    const res = await openai.chat.completions.create({
      messages,
      model: 'gpt-4o-mini',
      temperature: 0,
      tools: searchTool.functions.toolSpecs
    })
    const message = res.choices?.[0]?.message
    console.log(message?.content)
  }
}

await main()
```

--------------------------------

### Install Hono MCP dependencies

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/ts-mcp-hono.mdx

Install the required packages for Hono, MCP, and Zod using your preferred package manager.

```bash
npm add hono @hono/node-server @hono/mcp @modelcontextprotocol/sdk zod
```

```bash
pnpm add hono @hono/node-server @hono/mcp @modelcontextprotocol/sdk zod
```

```bash
bun add hono @hono/node-server @hono/mcp @modelcontextprotocol/sdk zod
```

```bash
yarn add hono @hono/node-server @hono/mcp @modelcontextprotocol/sdk zod
```

--------------------------------

### Configure Pricing Plans in agentic.config.ts

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/config/pricing.mdx

Examples of defining various pricing strategies including free, freemium with trials, and metered pay-as-you-go models.

```ts
import { defineConfig } from '@agentic/platform'

export default defineConfig({
  // ...
  pricingPlans: [
    {
      name: 'Free',
      slug: 'free',
      lineItems: [
        {
          slug: 'base',
          usageType: 'licensed',
          amount: 0
        }
      ]
    }
  ]
})
```

```ts
import { defaultFreePricingPlan, defineConfig } from '@agentic/platform'

export default defineConfig({
  // ...
  pricingPlans: [
    defaultFreePricingPlan,
    {
      name: 'Basic',
      slug: 'basic',
      trialPeriodDays: 7,
      lineItems: [
        {
          slug: 'base',
          usageType: 'licensed',
          amount: 499 // $4.99 USD
        }
      ]
    }
  ]
})
```

```ts
import { defineConfig } from '@agentic/platform'

export default defineConfig({
  // ...
  pricingPlans: [
    {
      name: 'Free',
      slug: 'free',
      lineItems: [
        {
          slug: 'base',
          usageType: 'licensed',
          amount: 0
        }
      ],
      // Free but limited to 20 requests per day
      rateLimit: {
        limit: 20,
        interval: '1d'
      }
    },
    {
      name: 'Pay-As-You-Go',
      slug: 'pay-as-you-go',
      lineItems: [
        {
          slug: 'requests',
          usageType: 'metered',
          billingScheme: 'tiered',
          tiersMode: 'volume',
          // $0.00467 USD per request up to 999 requests per month
          // then $0.00053 USD for unlimited further requests that month
          tiers: [
            {
              upTo: 999,
              unitAmount: 0.467
            },
            {
              upTo: 'inf',
              unitAmount: 0.053
            }
          ]
        }
      ],
      // Limit to 1000 requests per day
      rateLimit: {
        limit: 1000,
        interval: '1d'
      }
    }
  ]
})
```

--------------------------------

### Use Agentic Search Tool with Vercel AI SDK

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/ai-sdk.mdx

Example demonstrating how to use the Agentic search tool with the Vercel AI SDK and OpenAI. Requires API keys for OpenAI and Agentic.

```typescript
import 'dotenv/config'

import { createAISDKTools } from '@agentic/ai-sdk'
import { AgenticToolClient } from '@agentic/platform-tool-client'
import { createOpenAI } from '@ai-sdk/openai'
import { generateText } from 'ai'

async function main() {
  const searchTool = await AgenticToolClient.fromIdentifier('@agentic/search')
  const openai = createOpenAI({ compatibility: 'strict' })

  const result = await generateText({
    model: openai('gpt-4o-mini'),
    tools: createAISDKTools(searchTool),
    toolChoice: 'required',
    temperature: 0,
    system: 'You are a helpful assistant. Be as concise as possible.',
    prompt: 'What is the weather in San Francisco?'
  })

  console.log(JSON.stringify(result.toolResults[0], null, 2))
}

await main()
```

--------------------------------

### Create and Run a Weather Agent (TypeScript)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/mastra.mdx

Example of creating a Mastra agent that uses the @agentic/search tool and the OpenAI provider to answer weather-related questions. Ensure you have 'dotenv/config' imported and OPENAI_API_KEY set in your .env file.

```typescript
import 'dotenv/config'

import { createMastraTools } from '@agentic/mastra'
import { AgenticToolClient } from '@agentic/platform-tool-client'
import { openai } from '@ai-sdk/openai'
import { Agent } from '@mastra/core/agent'

async function main() {
  const searchTool = await AgenticToolClient.fromIdentifier('@agentic/search')

  const weatherAgent = new Agent({
    name: 'Weather Agent',
    instructions: 'You are a helpful assistant. Be as concise as possible.',
    model: openai('gpt-4o-mini'),
    tools: createMastraTools(searchTool)
  })

  const res = await weatherAgent.generate(
    'What is the weather in San Francisco?'
  )
  console.log(res.text)
}

await main()
```

--------------------------------

### Example Agentic Deployment Output

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/existing-mcp-server.mdx

This JSON object represents the output of a successful deployment. It includes details about the deployment's ID, creation timestamp, identifier, published status, associated tools, pricing plans, and gateway URLs.

```json
{
  "id": "depl_kf4c3o8efh2y84dp5iwsha95",
  "createdAt": "2025-06-28 05:38:34.960754+00",
  "updatedAt": "2025-06-28 05:38:34.960754+00",
  "identifier": "@dev/search@42ad78bf",
  "hash": "42ad78bf",
  "published": false,
  "description": "Official Google Search tool. Useful for finding up-to-date news and information about any topic.",
  "readme": "",
  "userId": "user_bhlpuiioipxilpuq7xaoh1ae",
  "projectId": "proj_rxs9jorlwolc3seq8enqgrgc",
  "tools": [
    {
      "name": "search",
      "description": "Uses Google Search to return the most relevant web pages for a given query. Useful for finding up-to-date news and information about any topic.",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "type": "object",
        "properties": {
          "num": {
            "type": "integer",
            "default": 5,
            "description": "Number of results to return"
          },
          "type": {
            "enum": [
              "search",
              "images",
              "videos",
              "places",
              "news",
              "shopping"
            ],
            "type": "string",
            "default": "search",
            "description": "Type of Google search to perform"
          },
          "query": {
            "type": "string",
            "description": "Search query"
          }
        },
        "required": ["query"],
        "additionalProperties": false
      },
      "outputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "type": "object",
        "properties": {
          "news": {},
          "images": {},
          "places": {},
          "videos": {},
          "results": {},
          "shopping": {},
          "answerBox": {},
          "knowledgeGraph": {}
        },
        "additionalProperties": false
      }
    }
  ],
  "toolConfigs": [
    {
      "name": "search",
      "cacheControl": "public, max-age=60, s-maxage=60 stale-while-revalidate=10"
    }
  ],
  "pricingPlans": [
    {
      "name": "Free",
      "slug": "free",
      "rateLimit": {
        "interval": 86400,
        "limit": 10,
        "mode": "approximate",
        "enabled": true
      },
      "lineItems": [
        {
          "slug": "requests",
          "usageType": "metered",
          "billingScheme": "per_unit",
          "unitAmount": 0
        }
      ]
    },
    {
      "name": "Standard",
      "slug": "standard",
      "interval": "month",
      "rateLimit": {
        "interval": 1,
        "limit": 100,
        "mode": "approximate",
        "enabled": true
      },
      "lineItems": [
        {
          "slug": "base",
          "usageType": "licensed",
          "amount": 1000
        },
        {
          "slug": "requests",
          "usageType": "metered",
          "billingScheme": "tiered",
          "tiersMode": "volume",
          "tiers": [
            {
              "unitAmount": 0,
              "upTo": 1000
            },
            {
              "unitAmount": 0.01,
              "upTo": 50000
            },
            {
              "unitAmount": 0.008,
              "upTo": 500000
            },
            {
              "unitAmount": 0.006,
              "upTo": 2500000
            },
            {
              "unitAmount": 0.005,
              "upTo": "inf"
            }
          ]
        }
      ]
    }
  ],
  "pricingIntervals": ["month"],
  "defaultRateLimit": {
    "interval": 60,
    "limit": 1000,
    "mode": "approximate",
    "enabled": true
  },
  "gatewayBaseUrl": "https://gateway.agentic.so/@dev/search@42ad78bf",
  "gatewayMcpUrl": "https://gateway.agentic.so/@dev/search@42ad78bf/mcp",
  "marketplaceUrl": "https://agentic.so/marketplace/projects/@dev/search",
  "adminUrl": "https://agentic.so/app/projects/@dev/search/deployments/42ad78bf"
}
```

--------------------------------

### Create a FastMCP server

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/ts-fastmcp.mdx

Initialize a FastMCP server instance and define a tool with Zod validation. The server must be started with the httpStream transport type.

```typescript
import { FastMCP } from 'fastmcp'
import { z } from 'zod' // Or any validation library that supports Standard Schema

const server = new FastMCP({
  name: 'Example',
  version: '0.0.1'
})

server.addTool({
  name: 'add',
  description: 'Add two numbers',
  parameters: z.object({
    a: z.number(),
    b: z.number()
  }),
  execute: async (args) => {
    return String(args.a + args.b)
  }
})

server.start({
  transportType: 'httpStream',
  httpStream: {
    port: 8080
  }
})
```

--------------------------------

### Implement ModelFetch MCP Server in TypeScript

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/ts-modelfetch.mdx

Implement your MCP server by defining tools, resources, and prompts. This example registers an 'add' tool that sums two numbers.

```typescript
import { McpServer } from '@modelcontextprotocol/sdk/server/mcp.js'
import { z } from 'zod'

const server = new McpServer({
  title: 'My MCP Server',
  name: 'my-mcp-server',
  version: '1.0.0'
})

server.registerTool(
  'add',
  {
    title: 'Add Numbers',
    description: 'Add two numbers together',
    inputSchema: {
      a: z.number(),
      b: z.number()
    }
  },
  ({ a, b }) => ({
    content: [
      {
        type: 'text',
        text: `${a} + ${b} = ${a + b}`
      }
    ]
  })
)

export default server
```

--------------------------------

### Build All Packages and Apps

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/CLAUDE.md

This command builds all packages and applications, excluding the website, for deployment or testing.

```bash
pnpm build
```

--------------------------------

### Navigate to Project Directory

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/ts-modelfetch.mdx

After creating the ModelFetch project, navigate to the project directory using this command.

```bash
cd my-mcp-server
```

--------------------------------

### Install @agentic/platform-validators

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/packages/validators/readme.md

Install the package using npm. This package is typically used internally by other Agentic packages.

```bash
npm i @agentic/platform-validators
```

--------------------------------

### View CLI Usage and Commands

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/packages/cli/readme.md

Displays the help menu and available commands for the Agentic CLI.

```sh
Usage: agentic [options] [command]

Options:
  -j, --json                                Print output as JSON
  -V, --version                             output the version number
  -h, --help                                display help for command

Commands:
  login|signin [options]                    Signs in to Agentic. If no credentials are provided, uses GitHub auth.
  signup [options]                          Creates a new account for Agentic. If no credentials are provided, uses
                                            GitHub auth.
  whoami                                    Displays info about the current user.
  logout|signout                            Signs the current user out.
  deploy [options]                          Creates a new deployment.
  publish [options] [deploymentIdentifier]  Publishes a deployment. Defaults to the most recent deployment for the
                                            project in the target directory. If a deployment identifier is provided, it
                                            will be used instead.
  get <deploymentIdentifier>                Gets details for a specific deployment.
  list|ls [options] [identifier]            Lists deployments.
  debug [options]                           Prints config for a local project.
  help [command]                            display help for command
```

--------------------------------

### Install Agentic Genkit dependencies

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/genkit.mdx

Install the required packages to use Agentic tools with Firebase Genkit.

```npm
npm install genkit @agentic/genkit @agentic/platform-tool-client
```

```pnpm
pnpm add genkit @agentic/genkit @agentic/platform-tool-client
```

```bun
bun add genkit @agentic/genkit @agentic/platform-tool-client
```

```yarn
yarn add genkit @agentic/genkit @agentic/platform-tool-client
```

--------------------------------

### Create a FastMCP server

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/py-fastmcp.mdx

Create a FastMCP server instance and define a tool function. Run the server using the Streamable HTTP transport.

```python
from fastmcp import FastMCP

mcp = FastMCP("Demo 🚀")

@mcp.tool
def add(a: int, b: int) -> int:
    """Add two numbers"""
    return a + b

if __name__ == "__main__":
    mcp.run(transport="http", host="127.0.0.1", port=8000, path="/mcp")
```

--------------------------------

### Log in or sign up with email and password

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/existing-mcp-server.mdx

Log in to an existing Agentic account or create a new one using email and password credentials.

```bash
agentic login -e <email> -p <password>
```

```bash
agentic signup -e <email> -p <password> -u <username>
```

--------------------------------

### Install @agentic/platform

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/packages/platform/readme.md

Install the @agentic/platform package using npm. This command is used to add the package to your project's dependencies.

```bash
npm i @agentic/platform
```

--------------------------------

### Create ModelFetch MCP Server

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/ts-modelfetch.mdx

Use this command to create a new ModelFetch MCP server project. Follow the interactive prompts to choose a runtime, language, and package manager.

```bash
npx -y create-modelfetch@latest
```

--------------------------------

### Basic OpenAPI Origin Configuration (TypeScript)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/config/examples.mdx

Configure the Agentic MCP gateway with an origin OpenAPI server. Requires `@agentic/platform` and a local OpenAPI specification file.

```typescript
import { defineConfig } from '@agentic/platform'

export default defineConfig({
  name: 'Basic OpenAPI Example',
  description:
    "This example shows how to configure Agentic's MCP gateway with an origin OpenAPI server.",
  origin: {
    type: 'openapi',
    url: 'https://jsonplaceholder.typicode.com',
    spec: './jsonplaceholder.json'
  }
})
```

--------------------------------

### Install Agentic CLI Globally

Source: https://context7.com/transitive-bullshit/chatgpt-api/llms.txt

Install the Agentic CLI globally using npm. This command is used for managing Agentic projects and deployments.

```bash
npm install -g @agentic/cli
```

--------------------------------

### Deploy Project with Agentic CLI

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/existing-mcp-server.mdx

Run this command from the directory containing your agentic configuration file to create a new immutable preview deployment. Subsequent changes can be deployed using the same command.

```bash
agentic deploy
```

--------------------------------

### GET /@username/tool-name/mcp

Source: https://context7.com/transitive-bullshit/chatgpt-api/llms.txt

Provides the MCP endpoint for configuring clients like Claude Desktop.

```APIDOC
## GET /@username/tool-name/mcp

### Description
Returns the MCP configuration endpoint for a specific tool, used to integrate Agentic tools into MCP-compliant clients.

### Method
GET

### Endpoint
https://gateway.agentic.so/@username/tool-name/mcp
```

--------------------------------

### Use Agentic tools with Genkit

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/marketplace/ts-sdks/genkit.mdx

Example implementation using the @agentic/search tool with Genkit and OpenAI.

```ts
import 'dotenv/config'

import { createGenkitTools } from '@agentic/genkit'
import { AgenticToolClient } from '@agentic/platform-tool-client'
import { genkit } from 'genkit'
import { gpt4oMini, openAI } from 'genkitx-openai'

async function main() {
  const searchTool = await AgenticToolClient.fromIdentifier('@agentic/search')

  const ai = genkit({
    plugins: [openAI()]
  })

  const result = await ai.generate({
    model: gpt4oMini,
    tools: createGenkitTools(ai, searchTool),
    system: 'You are a helpful assistant. Be as concise as possible.',
    prompt: 'What is the weather in San Francisco?'
  })

  console.log(result)
}

await main()
```

--------------------------------

### Publish an Agentic deployment

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/existing-mcp-server.mdx

Execute this command to make your deployment publicly available and enable Stripe subscriptions.

```bash
agentic publish
```

--------------------------------

### Disabled Rate Limit Configuration

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/config/rate-limits.mdx

An example of how to disable a rate limit by setting the 'enabled' property to false.

```ts
{
  enabled: false
}
```

--------------------------------

### HTTP Endpoint

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/todo.md

The HTTP endpoint allows for standard GET and POST requests to interact with chatbot tools.

```APIDOC
## HTTP Endpoint

### Description
This endpoint allows for standard HTTP requests (GET, POST) to interact with chatbot tools.

### Method
GET, POST

### Endpoint
`https://gateway.agentic.so/@{username}/{project-slug}/{tool-name}`
```

--------------------------------

### Get Details of a Specific Agentic Deployment

Source: https://context7.com/transitive-bullshit/chatgpt-api/llms.txt

Retrieve detailed information about a specific Agentic deployment using its username and project name.

```bash
agentic get @username/project-name
```

--------------------------------

### Deploy to Vercel

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/ts-xmcp.mdx

Deploy the pre-built application to Vercel for production. This command assumes the project has already been built for Vercel.

```bash
vercel deploy --prod --prebuilt
```

--------------------------------

### Create a Hono MCP Node.js server

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/ts-mcp-hono.mdx

Initialize a Hono application and register an MCP tool using the @hono/mcp middleware.

```typescript
import { StreamableHTTPTransport } from '@hono/mcp'
import { serve } from '@hono/node-server'
import { McpServer } from '@modelcontextprotocol/sdk/server/mcp.js'
import { Hono } from 'hono'
import { logger } from 'hono/logger'
import { z } from 'zod'

const app = new Hono()
app.use(logger())

const mcpServer = new McpServer({
  name: 'Example',
  version: '0.0.1'
})

mcpServer.registerTool(
  'add',
  {
    description: 'Add two numbers',
    inputSchema: z.object({
      a: z.number(),
      b: z.number()
    })
  },
  async (args) => {
    return {
      content: [
        {
          type: 'text',
          content: String(args.a + args.b)
        }
      ]
    }
  }
)

app.all('/mcp', async (c) => {
  const transport = new StreamableHTTPTransport()
  await mcpServer.connect(transport)
  return transport.handleRequest(c)
})

serve({ fetch: app.fetch, port: 8787 })
```

--------------------------------

### Incorrect ESM Import Example (with .ts extension)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/CLAUDE.md

Illustrates an incorrect ESM import that includes the `.ts` file extension. Omit extensions for module imports.

```javascript
import { Foo } from './foo.ts'
```

--------------------------------

### Build xmcp App for Vercel

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/docs/publishing/guides/ts-xmcp.mdx

Build the MCP server specifically for deployment on Vercel. This command prepares the project for Vercel's hosting environment.

```bash
xmcp build --vercel
```

--------------------------------

### List Agentic Deployments as JSON

Source: https://context7.com/transitive-bullshit/chatgpt-api/llms.txt

List all Agentic deployments and output the result in JSON format, suitable for scripting.

```bash
agentic list --json
```

--------------------------------

### Incorrect ESM Import with Types Example (with .js extension)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/CLAUDE.md

Shows an incorrect ESM import for types that includes the `.js` extension. File extensions should be omitted.

```javascript
import { type Route } from './types/root.js'
```

--------------------------------

### Incorrect ESM Import Example (with .js extension)

Source: https://github.com/transitive-bullshit/chatgpt-api/blob/main/CLAUDE.md

Highlights an incorrect way to import modules using ESM by including the `.js` file extension. This should be avoided.

```javascript
import { Foo } from './foo.js'
```