docs: improve docs

2024-08-17 06:03:17 -05:00 · 2024-08-17 06:03:17 -05:00 · e65a0e6820
commit e65a0e6820
--- a/docs/intro.mdx
+++ b/docs/intro.mdx
@ -1,6 +1,6 @@
 ---
-title: 'Agentic'
-description: 'AI agent stdlib that works with any LLM and TypeScript AI SDK.'
+title: Intro
+description: Agentic is an AI agent stdlib that works with any LLM and TypeScript AI SDK.
 ---

 Agentic is a **standard library of AI tools** which are **optimized for both normal TS-usage as well as LLM-based usage**.
--- a/docs/mint.json
+++ b/docs/mint.json
@ -35,20 +35,17 @@
  "navigation": [
    {
      "group": "Getting Started",
+      "pages": ["intro", "quickstart", "usage"]
+    },
+    {
+      "group": "AI SDKs",
      "pages": [
-        "intro",
-        "quickstart",
-        "usage",
-        {
-          "group": "AI SDKs",
-          "pages": [
-            "sdks/ai-sdk",
-            "sdks/langchain",
-            "sdks/llamaindex",
-            "sdks/genkit",
-            "sdks/dexter"
-          ]
-        }
+        "sdks/ai-sdk",
+        "sdks/langchain",
+        "sdks/llamaindex",
+        "sdks/genkit",
+        "sdks/dexter",
+        "sdks/openai"
      ]
    },
    {
--- a/docs/quickstart.mdx
+++ b/docs/quickstart.mdx
@ -1,9 +1,7 @@
 ---
-title: 'Quick Start'
+title: Quick Start
 ---

-## Install
-
 <Info>
  **Prerequisite**: All Agentic packages are [ESM
  only](https://gist.github.com/sindresorhus/a39789f98801d908bbc7ff3ecc99d99c)
@ -52,19 +50,19 @@ title: 'Quick Start'
      <Accordion title="Install individual AI tools">
        Docs for individual tools are available [here](/tools).

-        Here's an example of how to install the `@agentic/calculator` tool:
+        Here's an example of how to install the [Weather tool](/tools/weather):

        <CodeGroup>
          ```bash npm
-          npm install @agentic/calculator
+          npm install @agentic/weather
          ```

          ```bash yarn
-          yarn add @agentic/calculator
+          yarn add @agentic/weather
          ```

          ```bash pnpm
-          pnpm add @agentic/calculator
+          pnpm add @agentic/weather
          ```
        </CodeGroup>
      </Accordion>
@ -98,7 +96,7 @@ title: 'Quick Start'
          ```
        </CodeGroup>

-        See the [Agentic ⇒ Vercel AI SDK](/sdks/ai-sdk) docs for more details.
+        See the [AI SDK adapter docs](/sdks/ai-sdk) for usage details.
      </Accordion>

      <Accordion title="LangChain">
@ -116,7 +114,7 @@ title: 'Quick Start'
          ```
        </CodeGroup>

-        See the [Agentic ⇒ LangChain](/sdks/langchain) docs for more details.
+        See the [LangChain adapter docs](/sdks/langchain) for usage details.
      </Accordion>

      <Accordion title="LlamaIndex">
@ -134,7 +132,7 @@ title: 'Quick Start'
          ```
        </CodeGroup>

-        See the [Agentic ⇒ LlamaIndex](/sdks/llamaindex) docs for more details.
+        See the [LlamaIndex adapter docs](/sdks/llamaindex) for usage details.
      </Accordion>

      <Accordion title="Firebase Genkit">
@ -152,7 +150,7 @@ title: 'Quick Start'
          ```
        </CodeGroup>

-        See the [Agentic ⇒ Genkit](/sdks/genkit) docs for more details.
+        See the [Genkit adapter docs](/sdks/genkit) for usage details.
      </Accordion>

      <Accordion title="Dexa Dexter">
@ -170,7 +168,7 @@ title: 'Quick Start'
          ```
        </CodeGroup>

-        See the [Agentic ⇒ Dexter](/sdks/dexter) docs for more details.
+        See the [Dexter adapter docs](/sdks/dexter) for usage details.
      </Accordion>

      <Accordion title="OpenAI SDK">
@ -190,7 +188,7 @@ title: 'Quick Start'

        There's no need for a separate adapter with the OpenAI SDK since all agentic tools are compatible with OpenAI by default. You can use `AIFunctionSet.specs` for function calling or `AIFunctionSet.toolSpecs` for parallel tool calling.

-        See the [Agentic ⇒ OpenAI](/sdks/openai) docs for more details.
+        See the [OpenAI adapter docs](/sdks/openai) for usage details.
      </Accordion>

    </AccordionGroup>
@ -202,7 +200,3 @@ title: 'Quick Start'
  </Step>

 </Steps>
-
-## Usage
-
-TODO
--- a/docs/sdks/ai-sdk.mdx
+++ b/docs/sdks/ai-sdk.mdx
@ -0,0 +1,77 @@
+---
+title: Vercel AI SDK
+description: Agentic adapter for the Vercel AI SDK.
+---
+
+- package: `@agentic/ai-sdk`
+- exports: `function createAISDKTools`
+- [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/ai-sdk/src/ai-sdk.ts)
+- [Vercel AI SDK docs](https://sdk.vercel.ai)
+
+## Install
+
+<CodeGroup>
+```bash npm
+npm install @agentic/ai-sdk ai
+```
+
+```bash yarn
+yarn add @agentic/ai-sdk ai
+```
+
+```bash pnpm
+pnpm add @agentic/ai-sdk ai
+```
+
+</CodeGroup>
+
+## Usage
+
+This example also requires you to install `@ai-sdk/openai`.
+
+```ts
+import 'dotenv/config'
+
+import { createAISDKTools } from '@agentic/ai-sdk'
+import { WeatherClient } from '@agentic/weather'
+import { openai } from '@ai-sdk/openai'
+import { generateText } from 'ai'
+
+async function main() {
+  const weather = new WeatherClient()
+
+  const result = await generateText({
+    model: openai('gpt-4o-mini'),
+    tools: createAISDKTools(weather),
+    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(result.toolResults[0])
+}
+
+await main()
+```
+
+## Running this example
+
+<Info>
+  You'll need a free API key from [weatherapi.com](https://www.weatherapi.com)
+  to run this example. Store it in a local `.env` file as `WEATHER_API_KEY`.
+</Info>
+
+<Info>
+  You'll need an [OpenAI API key](https://platform.openai.com/docs/quickstart)
+  to run this example. Store it in a local `.env` file as `OPENAI_API_KEY`.
+</Info>
+
+```sh
+git clone git@github.com:transitive-bullshit/agentic.git
+cd agentic
+pnpm install
+echo 'WEATHER_API_KEY=your-key' >> .env
+echo 'OPENAI_API_KEY=your-key' >> .env
+npx tsx examples/ai-sdk/bin/weather.ts
+```
--- a/docs/sdks/dexter.mdx
+++ b/docs/sdks/dexter.mdx
@ -0,0 +1,75 @@
+---
+title: Dexter
+description: Agentic adapter for the Dexa Dexter SDK.
+---
+
+- package: `@agentic/dexter`
+- exports: `function createDexterFunctions`
+- [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/dexter/src/dexter.ts)
+- [Dexa Dexter SDK docs](https://dexter.dexa.ai)
+
+## Install
+
+<CodeGroup>
+```bash npm
+npm install @agentic/dexter @dexaai/dexter
+```
+
+```bash yarn
+yarn add @agentic/dexter @dexaai/dexter
+```
+
+```bash pnpm
+pnpm add @agentic/dexter @dexaai/dexter
+```
+
+</CodeGroup>
+
+## Usage
+
+```ts
+import 'dotenv/config'
+
+import { createDexterFunctions } from '@agentic/dexter'
+import { WeatherClient } from '@agentic/weather'
+import { ChatModel, createAIRunner } from '@dexaai/dexter'
+
+async function main() {
+  const weather = new WeatherClient()
+
+  const runner = createAIRunner({
+    chatModel: new ChatModel({
+      params: { model: 'gpt-4o-mini', temperature: 0 }
+      // debug: true
+    }),
+    functions: createDexterFunctions(weather),
+    systemMessage: 'You are a helpful assistant. Be as concise as possible.'
+  })
+
+  const result = await runner('What is the weather in San Francisco?')
+  console.log(result)
+}
+
+await main()
+```
+
+## Running this example
+
+<Info>
+  You'll need a free API key from [weatherapi.com](https://www.weatherapi.com)
+  to run this example. Store it in a local `.env` file as `WEATHER_API_KEY`.
+</Info>
+
+<Info>
+  You'll need an [OpenAI API key](https://platform.openai.com/docs/quickstart)
+  to run this example. Store it in a local `.env` file as `OPENAI_API_KEY`.
+</Info>
+
+```sh
+git clone git@github.com:transitive-bullshit/agentic.git
+cd agentic
+pnpm install
+echo 'WEATHER_API_KEY=your-key' >> .env
+echo 'OPENAI_API_KEY=your-key' >> .env
+npx tsx examples/dexter/bin/weather.ts
+```
--- a/docs/sdks/genkit.mdx
+++ b/docs/sdks/genkit.mdx
@ -0,0 +1,89 @@
+---
+title: Genkit
+description: Agentic adapter for the Firebase Genkit SDK.
+---
+
+- package: `@agentic/genkit`
+- exports: `function createGenkitTools`
+- [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/genkit/src/genkit.ts)
+- [Firebase Genkit docs](https://firebase.google.com/docs/genkit)
+
+## Install
+
+<CodeGroup>
+```bash npm
+npm install @agentic/genkit @genkit-ai/ai @genkit-ai/core
+```
+
+```bash yarn
+yarn add @agentic/genkit @genkit-ai/ai @genkit-ai/core
+```
+
+```bash pnpm
+pnpm add @agentic/genkit @genkit-ai/ai @genkit-ai/core
+```
+
+</CodeGroup>
+
+## Usage
+
+This example also requires you to install the [genkitx-openai](https://github.com/TheFireCo/genkit-plugins/tree/main/plugins/openai) package, which adds support for OpenAI to Genkit.
+
+```ts
+import 'dotenv/config'
+
+import { createGenkitTools } from '@agentic/genkit'
+import { WeatherClient } from '@agentic/stdlib'
+import { generate } from '@genkit-ai/ai'
+import { configureGenkit } from '@genkit-ai/core'
+import { gpt4oMini, openAI } from 'genkitx-openai'
+
+async function main() {
+  const weather = new WeatherClient()
+
+  configureGenkit({
+    plugins: [openAI()]
+  })
+
+  const result = await generate({
+    model: gpt4oMini,
+    tools: createGenkitTools(weather),
+    history: [
+      {
+        role: 'system',
+        content: [
+          {
+            text: 'You are a helpful assistant. Be as concise as possible.'
+          }
+        ]
+      }
+    ],
+    prompt: 'What is the weather in San Francisco?'
+  })
+
+  console.log(result)
+}
+
+await main()
+```
+
+## Running this example
+
+<Info>
+  You'll need a free API key from [weatherapi.com](https://www.weatherapi.com)
+  to run this example. Store it in a local `.env` file as `WEATHER_API_KEY`.
+</Info>
+
+<Info>
+  You'll need an [OpenAI API key](https://platform.openai.com/docs/quickstart)
+  to run this example. Store it in a local `.env` file as `OPENAI_API_KEY`.
+</Info>
+
+```sh
+git clone git@github.com:transitive-bullshit/agentic.git
+cd agentic
+pnpm install
+echo 'WEATHER_API_KEY=your-key' >> .env
+echo 'OPENAI_API_KEY=your-key' >> .env
+npx tsx examples/genkit/bin/weather.ts
+```
--- a/docs/sdks/langchain.mdx
+++ b/docs/sdks/langchain.mdx
@ -0,0 +1,87 @@
+---
+title: LangChain
+description: Agentic adapter for the LangChain JS SDK.
+---
+
+- package: `@agentic/langchain`
+- exports: `function createLangChainTools`
+- [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/langchain/src/langchain.ts)
+- [LangChain JS docs](https://js.langchain.com)
+
+## Install
+
+<CodeGroup>
+```bash npm
+npm install @agentic/langchain @langchain/core langchain
+```
+
+```bash yarn
+yarn add @agentic/langchain @langchain/core langchain
+```
+
+```bash pnpm
+pnpm add @agentic/langchain @langchain/core langchain
+```
+
+</CodeGroup>
+
+## Usage
+
+```ts
+import { createLangChainTools } from '@agentic/langchain'
+import { WeatherClient } from '@agentic/stdlib'
+import { ChatPromptTemplate } from '@langchain/core/prompts'
+import { ChatOpenAI } from '@langchain/openai'
+import { AgentExecutor, createToolCallingAgent } from 'langchain/agents'
+
+async function main() {
+  const weather = new WeatherClient()
+
+  const tools = createLangChainTools(weather)
+  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()
+```
+
+## Running this example
+
+<Info>
+  You'll need a free API key from [weatherapi.com](https://www.weatherapi.com)
+  to run this example. Store it in a local `.env` file as `WEATHER_API_KEY`.
+</Info>
+
+<Info>
+  You'll need an [OpenAI API key](https://platform.openai.com/docs/quickstart)
+  to run this example. Store it in a local `.env` file as `OPENAI_API_KEY`.
+</Info>
+
+```sh
+git clone git@github.com:transitive-bullshit/agentic.git
+cd agentic
+pnpm install
+echo 'WEATHER_API_KEY=your-key' >> .env
+echo 'OPENAI_API_KEY=your-key' >> .env
+npx tsx examples/langchain/bin/weather.ts
+```
--- a/docs/sdks/llamaindex.mdx
+++ b/docs/sdks/llamaindex.mdx
@ -0,0 +1,76 @@
+---
+title: LlamaIndex
+description: Agentic adapter for the LlamaIndex TS SDK.
+---
+
+- package: `@agentic/llamaindex`
+- exports: `function createLlamaIndexTools`
+- [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/llamaindex/src/llamaindex.ts)
+- [LlamaIndex TS docs](https://ts.llamaindex.ai)
+
+## Install
+
+<CodeGroup>
+```bash npm
+npm install @agentic/llamaindex llamaindex
+```
+
+```bash yarn
+yarn add @agentic/llamaindex llamaindex
+```
+
+```bash pnpm
+pnpm add @agentic/llamaindex llamaindex
+```
+
+</CodeGroup>
+
+## Usage
+
+```ts
+import 'dotenv/config'
+
+import { createLlamaIndexTools } from '@agentic/llamaindex'
+import { WeatherClient } from '@agentic/stdlib'
+import { OpenAI, OpenAIAgent } from 'llamaindex'
+
+async function main() {
+  const weather = new WeatherClient()
+
+  const tools = createLlamaIndexTools(weather)
+  const agent = new OpenAIAgent({
+    llm: new OpenAI({ model: 'gpt-4o-mini', temperature: 0 }),
+    systemPrompt: 'You are a helpful assistant. Be as concise as possible.',
+    tools
+  })
+
+  const response = await agent.chat({
+    message: 'What is the weather in San Francisco?'
+  })
+
+  console.log(response.message.content)
+}
+
+await main()
+```
+
+## Running this example
+
+<Info>
+  You'll need a free API key from [weatherapi.com](https://www.weatherapi.com)
+  to run this example. Store it in a local `.env` file as `WEATHER_API_KEY`.
+</Info>
+
+<Info>
+  You'll need an [OpenAI API key](https://platform.openai.com/docs/quickstart)
+  to run this example. Store it in a local `.env` file as `OPENAI_API_KEY`.
+</Info>
+
+```sh
+git clone git@github.com:transitive-bullshit/agentic.git
+cd agentic
+pnpm install
+echo 'WEATHER_API_KEY=your-key' >> .env
+echo 'OPENAI_API_KEY=your-key' >> .env
+npx tsx examples/llamaindex/bin/weather.ts
+```
--- a/docs/sdks/openai.mdx
+++ b/docs/sdks/openai.mdx
@ -0,0 +1,83 @@
+---
+title: OpenAI
+description: How to use Agentic with the OpenAI TS SDK directly.
+---
+
+<Note>
+  There's no need for an adapter with the OpenAI SDK since all agentic tools are
+  compatible with OpenAI by default. You can use `AIFunctionSet.specs` for
+  function calling or `AIFunctionSet.toolSpecs` for parallel tool calling.
+</Note>
+
+## Install
+
+<CodeGroup>
+```bash npm
+npm install @agentic/core @agentic/stdlib openai
+```
+
+```bash yarn
+yarn add @agentic/core @agentic/stdlib openai
+```
+
+```bash pnpm
+pnpm add @agentic/core @agentic/stdlib openai
+```
+
+</CodeGroup>
+
+## Usage
+
+```ts
+import { WeatherClient } from '@agentic/stdlib'
+import OpenAI from 'openai'
+
+const weather = new WeatherClient()
+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?' }
+]
+
+const res = await openai.chat.completions.create({
+  messages,
+  model: 'gpt-4o-mini',
+  temperature: 0,
+  tools: weather.functions.toolSpecs,
+  tool_choice: 'required'
+})
+const message = res.choices[0]?.message!
+console.log(JSON.stringify(message, null, 2))
+assert(message.tool_calls?.[0]?.function?.name === 'get_current_weather')
+
+const fn = weather.functions.get('get_current_weather')!
+
+const toolParams = message.tool_calls[0].function.arguments
+const toolResult = await fn(toolParams)
+console.log(JSON.stringify(toolResult, null, 2))
+```
+
+## Running this example
+
+<Info>
+  You'll need a free API key from [weatherapi.com](https://www.weatherapi.com)
+  to run this example. Store it in a local `.env` file as `WEATHER_API_KEY`.
+</Info>
+
+<Info>
+  You'll need an [OpenAI API key](https://platform.openai.com/docs/quickstart)
+  to run this example. Store it in a local `.env` file as `OPENAI_API_KEY`.
+</Info>
+
+```sh
+git clone git@github.com:transitive-bullshit/agentic.git
+cd agentic
+pnpm install
+echo 'WEATHER_API_KEY=your-key' >> .env
+echo 'OPENAI_API_KEY=your-key' >> .env
+npx tsx examples/openai/bin/weather.ts
+```
--- a/docs/tools/bing.mdx
+++ b/docs/tools/bing.mdx
@ -7,7 +7,7 @@ description: Bing web search API client.
 - exports: `class BingClient`, `namespace bing`
 - env vars: `BING_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/bing/src/bing-client.ts)
- [api docs](https://www.microsoft.com/en-us/bing/apis/bing-web-search-api)
+- [bing web search docs](https://www.microsoft.com/en-us/bing/apis/bing-web-search-api)

 ## Install

--- a/docs/tools/calculator.mdx
+++ b/docs/tools/calculator.mdx
@ -7,7 +7,7 @@ description: A simple calculator tool.
 - exports: `function calculator`
 - env vars: _none_
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/calculator/src/calculator.ts)
- [api docs](https://mathjs.org)
+- [mathjs docs](https://mathjs.org)

 ## Install

--- a/docs/tools/clearbit.mdx
+++ b/docs/tools/clearbit.mdx
@ -7,7 +7,7 @@ description: Resolving and enriching people and company data.
 - exports: `class ClearbitClient`, `namespace clearbit`
 - env vars: `CLEARBIT_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/clearbit/src/clearbit-client.ts)
- [api docs](https://dashboard.clearbit.com/docs)
+- [clearbit api docs](https://dashboard.clearbit.com/docs)

 ## Install

--- a/docs/tools/diffbot.mdx
+++ b/docs/tools/diffbot.mdx
@ -9,7 +9,7 @@ Diffbot provides web page classification and scraping. It also provides access t
 - exports: `class DiffbotClient`, `namespace diffbot`
 - env vars: `DIFFBOT_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/diffbot/src/diffbot-client.ts)
- [api docs](https://docs.diffbot.com)
+- [diffbot api docs](https://docs.diffbot.com)

 ## Install

--- a/docs/tools/e2b.mdx
+++ b/docs/tools/e2b.mdx
@ -9,7 +9,7 @@ Hosted Python code intrepreter sandbox which is really useful for data analysis,
 - exports: `function e2b`
 - env vars: `E2B_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/e2b/src/e2b.ts)
- [api docs](https://e2b.dev)
+- [e2b api docs](https://e2b.dev)

 ## Install

--- a/docs/tools/exa.mdx
+++ b/docs/tools/exa.mdx
@ -7,7 +7,7 @@ description: Web search tailored for LLMs.
 - exports: `class ExaClient`, `namespace exa`
 - env vars: `EXA_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/exa/src/exa-client.ts)
- [api docs](https://docs.exa.ai)
+- [exa api docs](https://docs.exa.ai)

 ## Install

--- a/docs/tools/firecrawl.mdx
+++ b/docs/tools/firecrawl.mdx
@ -9,7 +9,7 @@ Turn websites into LLM-ready data. Crawl and convert any website into clean mark
 - exports: `class FirecrawlClient`, `namespace firecrawl`
 - env vars: `FIRECRAWL_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/firecrawl/src/firecrawl-client.ts)
- [api docs](https://www.firecrawl.dev)
+- [firecrawl api docs](https://www.firecrawl.dev)

 ## Install

--- a/docs/tools/hacker-news.mdx
+++ b/docs/tools/hacker-news.mdx
@ -9,7 +9,7 @@ Note that the [HN Algolia API](https://hn.algolia.com/api) seems to no longer be
 - exports: `class HackerNewsClient`, `namespace hackernews`
 - env vars: `HACKER_NEWS_API_USER_AGENT` (_optional_)
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/hacker-news/src/hacker-news-client.ts)
- [api docs](https://github.com/HackerNews/API)
+- [HN api docs](https://github.com/HackerNews/API)

 ## Install

--- a/docs/tools/hunter.mdx
+++ b/docs/tools/hunter.mdx
@ -7,7 +7,7 @@ description: Email finder, verifier, and enrichment.
 - exports: `class HunterClient`, `namespace hunter`
 - env vars: `HUNTER_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/hunter/src/hunter-client.ts)
- [api docs]()
+- [hunter api docs](https://hunter.io/api)

 ## Install

--- a/docs/tools/jina.mdx
+++ b/docs/tools/jina.mdx
@ -9,7 +9,7 @@ LLM-friendly URL reader and search client by [Jina AI](https://jina.ai/reader) w
 - exports: `class JinaClient`, `namespace jina`
 - env vars: `JINA_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/jina/src/jina-client.ts)
- [api docs](https://jina.ai/reader)
+- [jina api docs](https://jina.ai/reader)

 ## Install

--- a/docs/tools/midjourney.mdx
+++ b/docs/tools/midjourney.mdx
@ -7,7 +7,7 @@ description: Unofficial Midjourney API client for generative images.
 - exports: `class MidjourneyClient`, `namespace midjourney`
 - env vars: `MIDJOURNEY_IMAGINE_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/midjourney/src/midjourney-client.ts)
- [api docs](https://www.imagineapi.dev)
+- [imagine api docs](https://www.imagineapi.dev)

 ## Install

--- a/docs/tools/novu.mdx
+++ b/docs/tools/novu.mdx
@ -9,7 +9,7 @@ The [Novu API](https://novu.co) provides a router for sending notifications acro
 - exports: `class NovuClient`, `namespace novu`
 - env vars: `NOVU_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/novu/src/novu-client.ts)
- [api docs](https://novu.co)
+- [novu api docs](https://novu.co)

 ## Install

--- a/docs/tools/people-data-labs.mdx
+++ b/docs/tools/people-data-labs.mdx
@ -12,7 +12,7 @@ description: People & company data enrichment.
 - exports: `class PeopleDataLabsClient`, `namespace peopledatalabs`
 - env vars: `PEOPLE_DATA_LABS_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/people-data-labs/src/people-data-labs-client.ts)
- [api docs](https://www.peopledatalabs.com)
+- [people data labs api docs](https://www.peopledatalabs.com)

 ## Install

--- a/docs/tools/perigon.mdx
+++ b/docs/tools/perigon.mdx
@ -13,7 +13,7 @@ Real-time news API and web content data from 140,000+ sources. Structured and en
 - exports: `class PerigonClient`, `namespace perigon`
 - env vars: `PERIGON_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/perigon/src/perigon-client.ts)
- [api docs](https://www.goperigon.com/products/news-api)
+- [perigon api docs](https://www.goperigon.com/products/news-api)

 ## Install

--- a/docs/tools/polygon.mdx
+++ b/docs/tools/polygon.mdx
@ -7,7 +7,7 @@ description: Stock market and company financial data.
 - exports: `class PolygonClient`, `namespace polygon`
 - env vars: `POLYGON_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/polygon/src/polygon-client.ts)
- [api docs](https://polygon.io/docs)
+- [polygon api docs](https://polygon.io/docs)

 ## Install

--- a/docs/tools/predict-leads.mdx
+++ b/docs/tools/predict-leads.mdx
@ -9,7 +9,7 @@ In-depth company data, including signals like fundraising announcemnts, hiring i
 - exports: `class PredictLeadsClient`, `namespace predictleads`
 - env vars: `PREDICT_LEADS_API_KEY`, `PREDICT_LEADS_API_TOKEN`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/predict-leads/src/predict-leads-client.ts)
- [api docs]()
+- [precit leads api docs](https://docs.predictleads.com/v3)

 ## Install

--- a/docs/tools/proxycurl.mdx
+++ b/docs/tools/proxycurl.mdx
@ -7,7 +7,7 @@ description: People and company data from LinkedIn & Crunchbase.
 - exports: `class ProxycurlClient`, `namespace proxycurl`
 - env vars: `PROXYCURL_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/proxycurl/src/proxycurl-client.ts)
- [api docs](https://nubela.co/proxycurl)
+- [proxycurl api docs](https://nubela.co/proxycurl)

 ## Install

--- a/docs/tools/searxng.mdx
+++ b/docs/tools/searxng.mdx
@ -27,7 +27,7 @@ The most important search engines are:
 - exports: `class SearxngClient`, `namespace searxng`
 - env vars: `SEARXNG_API_BASE_URL`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/searxng/src/searxng-client.ts)
- [api docs](https://docs.searxng.org)
+- [searxng api docs](https://docs.searxng.org)

 ## Install

--- a/docs/tools/serpapi.mdx
+++ b/docs/tools/serpapi.mdx
@ -7,7 +7,7 @@ description: Lightweight wrapper around SerpAPI for Google search.
 - exports: `class SerpAPIClient`, `namespace serpapi`
 - env vars: `SERPAPI_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/serpapi/src/serpapi-client.ts)
- [api docs](https://serpapi.com/search-api)
+- [serpapi api docs](https://serpapi.com/search-api)

 ## Install

--- a/docs/tools/serper.mdx
+++ b/docs/tools/serper.mdx
@ -7,7 +7,7 @@ description: Lightweight wrapper around Serper for Google search.
 - exports: `class SerperClient`, `namespace serper`
 - env vars: `SERPER_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/serper/src/serper-client.ts)
- [api docs](https://serper.dev)
+- [serper api docs](https://serper.dev)

 ## Install

--- a/docs/tools/slack.mdx
+++ b/docs/tools/slack.mdx
@ -7,7 +7,7 @@ description: Minimal Slack API client for sending and receiving Slack messages.
 - exports: `class SlackClient`, `namespace slack`
 - env vars: `SLACK_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/slack/src/slack-client.ts)
- [api docs]()
+- [slack api docs](https://api.slack.com/docs)

 ## Install

--- a/docs/tools/social-data.mdx
+++ b/docs/tools/social-data.mdx
@ -7,7 +7,7 @@ description: Unofficial Twitter / X client (readonly) which is much cheaper than
 - exports: `class SocialDataClient`, `namespace socialdata`
 - env vars: `SOCIAL_DATA_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/social-data/src/social-data-client.ts)
- [api docs](https://socialdata.tools)
+- [social data tools api docs](https://socialdata.tools)

 ## Install

--- a/docs/tools/tavily.mdx
+++ b/docs/tools/tavily.mdx
@ -7,7 +7,7 @@ description: Web search API tailored for LLMs.
 - exports: `class TavilyClient`, `namespace tavily`
 - env vars: `BING_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/tavily/src/tavily-client.ts)
- [api docs](https://tavily.com)
+- [tavily api docs](https://tavily.com)

 ## Install

--- a/docs/tools/twilio.mdx
+++ b/docs/tools/twilio.mdx
@ -7,7 +7,7 @@ description: Twilio conversation API to send and receive SMS messages.
 - exports: `class TwilioClient`, `namespace twilio`
 - env vars: `TWILIO_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/twilio/src/twilio-client.ts)
- [api docs](https://www.twilio.com/docs/conversations/api)
+- [twilio api docs](https://www.twilio.com/docs/conversations/api)

 ## Install

--- a/docs/tools/twitter.mdx
+++ b/docs/tools/twitter.mdx
@ -9,7 +9,7 @@ Basic Twitter API methods for fetching users, tweets, and searching recent tweet
 - exports: `class TwitterClient`, `namespace twitter`
 - env vars: `TWITTER_API_KEY`, `TWITTER_API_PLAN`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/twitter/src/twitter-client.ts)
- [api docs](https://developer.twitter.com/en/docs/twitter-api)
+- [twitter api docs](https://developer.twitter.com/en/docs/twitter-api)

 ## Install

--- a/docs/tools/weather.mdx
+++ b/docs/tools/weather.mdx
@ -7,7 +7,7 @@ description: Simple Weather API client for accessing weather data based on locat
 - exports: `class WeatherClient`, `namespace weather`
 - env vars: `WEATHER_API_KEY`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/weather/src/weather-client.ts)
- [api docs](https://www.weatherapi.com)
+- [weatherapi.com api docs](https://www.weatherapi.com)

 ## Install

--- a/docs/tools/wikidata.mdx
+++ b/docs/tools/wikidata.mdx
@ -7,7 +7,7 @@ description: Basic Wikidata client.
 - exports: `class WikidataClient`, `namespace wikidata`
 - env vars: `WIKIDATA_API_USER_AGENT` (_optional_)
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/wikidata/src/wikidata-client.ts)
- [api docs](https://github.com/maxlath/wikibase-sdk)
+- [wikibase api docs](https://github.com/maxlath/wikibase-sdk)

 ## Install

--- a/docs/tools/wikipedia.mdx
+++ b/docs/tools/wikipedia.mdx
@ -7,7 +7,7 @@ description: Wikipedia apage search and summary API.
 - exports: `class WikipediaClient`, `namespace wikipedia`
 - env vars: `WIKIPEDIA_API_USER_AGENT` (_optional_)
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/wikipedia/src/wikipedia-client.ts)
- [api docs](https://www.mediawiki.org/wiki/API)
+- [wikipedia api docs](https://www.mediawiki.org/wiki/API)

 ## Install

--- a/docs/tools/wolfram-alpha.mdx
+++ b/docs/tools/wolfram-alpha.mdx
@ -7,7 +7,7 @@ description: Wolfram Alpha LLM API client for answering computational, mathemati
 - exports: `class WolframAlphaClient`, `namespace wolframalpha`
 - env vars: `WOLFRAM_APP_ID`
 - [source](https://github.com/transitive-bullshit/agentic/blob/main/packages/wolfram-alpha/src/wolfram-alpha-client.ts)
- [api docs](https://products.wolframalpha.com/llm-api/documentation)
+- [wolfram alpha api docs](https://products.wolframalpha.com/llm-api/documentation)

 ## Install

--- a/docs/usage.mdx
+++ b/docs/usage.mdx
@ -0,0 +1,124 @@
+---
+title: Usage
+---
+
+## AI SDKs
+
+<CardGroup cols={2}>
+<Card title='Vercel AI SDK' href='/sdks/ai-sdk'>
+  Using Agentic with the Vercel AI SDK.
+</Card>
+
+<Card title='LangChain' href='/sdks/langchain'>
+  Using Agentic with LangChain.
+</Card>
+
+<Card title='LlamaIndex' href='/sdks/llamaindex'>
+  Using Agentic with LlamaIndex.
+</Card>
+
+<Card title='Firebase Genkit' href='/sdks/genkit'>
+  Using Agentic with Genkit.
+</Card>
+
+<Card title='Dexa Dexter' href='/sdks/dexter'>
+  Using Agentic with Dexter.
+</Card>
+
+<Card title='OpenAI' href='/sdks/openai'>
+  Using Agentic with OpenAI directly.
+</Card>
+</CardGroup>
+
+## Tools
+
+### TS Tool Usage
+
+Agentic clients like `WeatherClient` can be used as normal TS classes:
+
+```ts
+import { WeatherClient } from '@agentic/stdlib'
+
+// Requires `process.env.WEATHER_API_KEY` (free from weatherapi.com)
+const weather = new WeatherClient()
+
+const result = await weather.getCurrentWeather({
+  q: 'San Francisco'
+})
+console.log(result)
+```
+
+### LLM Tool Usage
+
+Or you can use these clients as **LLM-based tools** where the LLM decides when and how to invoke the underlying functions for you.
+
+This works across all of the major AI SDKs via adapters. Here's an example using [Vercel's AI SDK](https://github.com/vercel/ai):
+
+```ts
+// sdk-specific imports
+import { openai } from '@ai-sdk/openai'
+import { generateText } from 'ai'
+import { createAISDKTools } from '@agentic/ai-sdk'
+
+// sdk-agnostic imports
+import { WeatherClient } from '@agentic/stdlib'
+
+const weather = new WeatherClient()
+
+const result = await generateText({
+  model: openai('gpt-4o-mini'),
+  // this is the key line which uses the `@agentic/ai-sdk` adapter
+  tools: createAISDKTools(weather),
+  toolChoice: 'required',
+  prompt: 'What is the weather in San Francisco?'
+})
+
+console.log(result.toolResults[0])
+```
+
+You can use our standard library of thoroughly tested AI functions with your favorite AI SDK – without having to write any glue code!
+
+### Multiple Tool Usage via AIFunctionLike
+
+Here's a slightly more complex example which uses multiple clients and selects a subset of their functions using the `AIFunctionSet.pick` method:
+
+```ts
+// sdk-specific imports
+import { ChatModel, createAIRunner } from '@dexaai/dexter'
+import { createDexterFunctions } from '@agentic/dexter'
+
+// sdk-agnostic imports
+import { PerigonClient, SerperClient } from '@agentic/stdlib'
+
+async function main() {
+  // Perigon is a news API and Serper is a Google search API
+  const perigon = new PerigonClient()
+  const serper = new SerperClient()
+
+  const runner = createAIRunner({
+    chatModel: new ChatModel({
+      params: { model: 'gpt-4o-mini', temperature: 0 }
+    }),
+    functions: createDexterFunctions(
+      perigon.functions.pick('search_news_stories'),
+      serper
+    ),
+    systemMessage: 'You are a helpful assistant. Be as concise as possible.'
+  })
+
+  const result = await runner(
+    'Summarize the latest news stories about the upcoming US election.'
+  )
+  console.log(result)
+}
+```
+
+Here we've exposed 2 functions to the LLM, `search_news_stories` (which comes from the `PerigonClient.searchStories` method) and `serper_google_search` (which implicitly comes from the `SerperClient.search` method).
+
+All of the SDK adapters like `createDexterFunctions` accept very flexible `AIFunctionLike` objects, which include:
+
+- `AIFunctionSet` - Sets of AI functions (like `perigon.functions.pick('search_news_stories')` or `perigon.functions` or `serper.functions`)
+- `AIFunctionsProvider` - Client classes which expose an `AIFunctionSet` via the `.functions` property (like `perigon` or `serper`)
+- `AIFunction` - Individual functions (like `perigon.functions.get('search_news_stories')` or `serper.functions.get('serper_google_search')` or AI functions created directly via the `createAIFunction` utility function)
+
+You can pass as many of these `AIFunctionLike` objects as you'd like and you can manipulate them as `AIFunctionSet` sets via `.pick`, `.omit`, `.get`, `.map`, etc.
--- a/package.json
+++ b/package.json
@ -41,7 +41,6 @@
    "eslint": "^8.57.0",
    "husky": "^9.1.4",
    "lint-staged": "^15.2.9",
-    "mintlify": "^4.0.203",
    "npm-run-all2": "^6.2.2",
    "only-allow": "^1.2.1",
    "prettier": "^3.3.3",
--- a/pnpm-lock.yaml
+++ b/pnpm-lock.yaml
--- a/readme.md
+++ b/readme.md
@ -150,7 +150,7 @@ import { WeatherClient } from '@agentic/weather'

 ### AI SDKs

-To use Agentic with one of the supported AI SDKs, you'll also need to install its (_really lightweight_) adapter package.
+To use Agentic with one of the supported AI SDKs, you'll also need to install its corresponding adapter package.

 #### Vercel AI SDk