bitpro가 원문에서 글자만 모은 내용이에요. 표·그림·수식은 빠지거나 깨질 수 있어요. 전체 13쪽 중 5쪽
ng" alt=""><figcaption></figcaption></figure>
**Step 2-3: Task Analysis & MCP Selection**
AWESOME has selected two MCPs (**DexScreener** and **X**) to accomplish your task.<br>
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/687214368a27983809abeafd_AD_4nXfkIyadLuwUVi80Qx638UMwKZR-bc-0ZJ37KukOtul9guiftFRzUXrGckVGBqQ1hN3LBOFuAtpb8UJLcjURyiOQ6LW621_ABRrBBSzL0zN4H0nRFsYgs40wQ1V_mgpIVu-WLNag7w.png" alt=""><figcaption></figcaption></figure>
**Step 4: Workflow Suggestion**
AWESOME suggests researching memecoin data on DexScreener first, followed by a summarized post on X.\
*Note, X MCP requires authentication, so the user is asked to fill in Twitter API keys and other credentials.*<br>
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/687214368a27983809abeb0e_AD_4nXfs4D_orBuaDucDWWj-epqk8fT52W_s5lrxtVvrr5Mwu1qYig_vubtSOGiHl82kzzSzz-_MSXJoO5B8DMjAlDjV8d8_kabc0IuwhKTciyH97_psRdf0luHhFW0xOis8h92gWYI0UQ.png" alt=""><figcaption></figcaption></figure>
**Step 5: Execution & Output**
AWESOME builds and executes the workflow, and delivers the tweet on the user's X.<br>
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/687214368a27983809abeb0b_AD_4nXcWYKTvEJL_qBPYK_DRcgtSxaAZWMWd6Nf7tn0cULO6GuRD-45nuADIJ8KmIBa2x800uVpSpvqqbs0V7flSEBcjmTQc-9RzmYA4aDO4h1lRkBG-K_SbAJ6pmkvNFLaS_tZTvTCSZA.png" alt=""><figcaption></figcaption></figure>
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/687214358a27983809abeae7_AD_4nXcpdQguQtE44RjX7U5-QdcodalBd_Kh09MS14YMPueGwE3-yf8MlXfD2mRuDcm-xOX-r86D-3OsT-EhS4y6BdCIdBODZnvLusKCY4EFpNCvWNbU2L1bxWjVRtmaX1n2sPw2L7ROag.png" alt=""><figcaption></figcaption></figure>
**Case 2: GitHub Issue Research + Documentation**
**Step 1: Describe Your Task**
“Help me go through and analyze all GitHub issues related to ElizaOS (<https://github.com/elizaOS/eliza>), and summarize in my Notion.”<br>
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/687214368a27983809abeb05_AD_4nXcGGBNVOTgZkGAiZy9jNu7xV9pusSNGBg_1Atr8QVPhqwAuCS7AqNvVpR2CJLaiAOVPNs_DLfJgfbSwV1LYtAUgQvRUn20r5mss2gr-DmkdXqiM3A7y5cUIA_ilyIZeHLSGiQj2.png" alt=""><figcaption></figcaption></figure>
**Step 2-3: Task Analysis & MCP Selection**
AWESOME has selected two MCPs (**GitHub** and **Notion**) to accomplish your task.<br>
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/687214368a27983809abeb14_AD_4nXeQiX_WKdXRAKdNH76i_2epNTIXsn4LrIV-M__1TIlE3WkPGyvxD1pQH4a_MusQgbcc41Xw4YrT_o2pVJVGVJ8h1H7xgcs-uaLjG0eY8xIgK-rp3b1V0tVR5gqeCujHAh0CbtBUcQ.png" alt=""><figcaption></figcaption></figure>
**Step 4: Workflow Suggestion**
AWESOME suggests researching issues on GitHub first, followed by documentation on Notion.\
*Note, both GitHub and Notion MCPs require authentication, so the user is asked to fill in relevant credentials.*<br>
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/687214368a27983809abeb11_AD_4nXdun-jTTF_ttkiG9YZc9Nwy1BgyigypSPf0K8Xzc_70UcSK7KJfIkcx_uEG6RLn0oHNhQYCyQj8bixS07GOkiLNmdPWyt1USRSqpyi0vG99F762PPtogIWg8ZyI_QtuR-qgOTluWw.png" alt=""><figcaption></figcaption></figure>
**Step 5: Execution & Output**
AWESOME builds and executes the workflow, and delivers the documentation on the user's Notion.<br>
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/687214368a27983809abeafa_AD_4nXc2cOJWV9RxYmyuNYysmFjJuiwNw_862nVAqdVej_NV2o2aIVydEJACV9VmWIOkO8GRU0ivxtlq6N6nn5zZlVRslmG57X1zHsFMLfkVyaq4TyJodUIsChDprKlRYwLf6nOZEyX-_Q.png" alt=""><figcaption></figcaption></figure>
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/687214368a27983809abeb08_AD_4nXf0cItekxgv8St8mMg358ydZWtVrtxevkczkyD2tewwvByy88yfnoP8tx90pF9kgy43WRS0MunNkcnNKnVieGKKEFM7WCPV2SuVZPk-mb1RbGGxQVjEittonMP0tbSfYkqHXetOaA.png" alt=""><figcaption></figcaption></figure>
AWESOME is your first look at what it means to turn natural language into autonomous action — across dozens of protocols and agents.
Ready to build? Start your first Task now on <https://mcp.awenetwork.ai>
# Agent Marketplace
We’re excited to announce the launch of one of AWESOME’s most powerful features yet — the **Agent Marketplace**. This upgrade lets you go beyond individual Tasks and start composing, saving, and sharing full **MCP-powered agents**.
Whether you’re building internal tools, community-facing bots, or experimental AI workflows, you can now **package your logic into reusable agents** — and run them with a single click.
## **Save & Publish MCP Workflows as Agents**
Every workflow you build through **AWE Cloud (Tasks)** can now be saved as an **Agent** — a persistent, customizable unit that remembers its instructions, toolset, and workflow logic.
You can choose to:
* **Save as Private Agent** – for personal or internal use
* **Publish as Public Agent** – so anyone can discover and run it
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/6878ef919465eba9e07091ad_1280X1280.PNG" alt=""><figcaption></figcaption></figure>
Each agent retains:
* Natural language task description
* Selected MCPs
* Workflow logic (including sequence and decision structure)
This lets you turn your most powerful Tasks into reusable autonomous units — no code required.
## **Agent Marketplace: Discover and Reuse**
To help kickstart the ecosystem, we’re launching with an initial set of **6 showcase agents**, all designed as **simple, single-MCP agents**. These are meant to demonstrate the core capabilities and serve as templates for your own ideas:
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/6878efe392233d77362e4ec3_Screenshot%202025-07-17%20at%206.12.21%E2%80%AFPM.png" alt=""><figcaption></figcaption></figure>
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/6878efec6fcbad04a9d3aa48_1280X1280%20(1).PNG" alt=""><figcaption></figcaption></figure>
You can run these agents directly, fork and remix them, or use them as blueprints for your multi-MCP builds.
## **Run an Agent with a Single Click**
Running an agent is as simple as clicking a button.
Let’s try it with the **DexScreener Agent**:
**Task:**
“Identify the top 3 meme coins launched over the last 24 hours.”
Just open the agent, click **“Run Agent”**, and let AWESOME handle the rest:
1. It uses the DexScreener MCP to fetch token data
2. Filters results based on category and launch time
3. Sorts by volume or price spike
4. Outputs a clean summary — instantly
You can customize the task input, tweak logic, or duplicate
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/6878f01906361aef0deb9770_4611ecc6-d817-46c7-b808-8281c0320b12.png" alt=""><figcaption></figcaption></figure>
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/6878f02292233d77362e6eeb_2b16bbe5-1d91-43c0-afcb-5310684fa846.png" alt=""><figcaption></figcaption></figure>
<figure><img src="https://cdn.prod.website-files.com/67c1ac10b6107f1ae02b8eae/6878f02c901d900dfe919206_8567d343-f8d9-437e-949a-37c8baa61e79.png" alt=""><figcaption></figcaption></figure>
<br>
# AWESOME X Agent
We’re excited to introduce AWESOME\_X\_Agent, a fully automated agent built with the AWESOME platform that empowers users to crawl, analyze, and post tweets or threads using natural language commands.Whether you're summarizing narratives, responding to market trends, or scheduling posts from research insights — this agent gives you a seamless way to engage with X (formerly Twitter) through the AWESOME workflow interface.
<figure><img src="https://csgvlxe6fyz7.sg.larksuite.com/space/api/box/stream/download/asynccode/?code=YjdjMDE2ZmQ0ZjhhYWFjZDdhYjgxMTJiMWMwNWNlOTBfb3lTOFAzcFFPMmpWbmRSaG9HWG5jdUxNaW1NMWJvdkdfVG9rZW46Q3g2RWJiakoxb3ZsQjN4WGpnaWxpWnhvZ0ZjXzE3NTQ3MzYxMDI6MTc1NDczOTcwMl9WNA" alt=""><figcaption></figcaption></figure>
## Workflow Detail
### MCPs Involved
* [Twitter-Client-MCP](https://github.com/mzkrasner/twitter-client-mcp): Enables reading public tweets and threads by ID or username.
* [X-MCP](https://github.com/vidhupv/x-mcp): Enables posting tweets, replies, and threads to your X account.
### Tools Involved
[Twitter-Client-MCP](https://github.com/mzkrasner/twitter-client-mcp)
* Get a Twitter profile by username
* Get the authenticated user's profile
* Get a specific tweet by ID
* Get tweets from a Twitter user
* Like a tweet
* Retweet a tweet
* Search for tweets
* Search for Twitter profiles
* Get a list of users following a Twitter user
* Get a list of users that a Twitter user is following
* Follow a Twitter user
[X-MCP](https://github.com/vidhupv/x-mcp)
* Post a new tweet
* Post a new thread
* Show draft
* Publish draft
* Delete draft
***
## Potential Use Cases
The core power of AWESOME\_X\_Agent lies in its ability to crawl tweet content and transform it into actionable posts.Here’s what you can use it for:
* 🔍 Summarize crypto Twitter threads into digestible recaps or TL;DRs
* 🧵 Generate automated threads from token market data, project updates, or daily recaps
* 📢 Convert project info or research into social-ready content
* 📬 Pull updates from tracked accounts and auto-summarize for reposting
* 🧠 Run social listening agents that track sentiment or trend shifts and respond accordingly
This agent can operate standalone or as part of larger multi-agent workflows.
***
### Illustrative Use Case:
One powerful use case is running a multi-account summarizer:
> Help me go through the most recent tweet of each of the below 9 X accounts:[@S4mmyEth](https://x.com/@S4mmyEth)[@Senti\_\_23](https://x.com/@Senti__23)[@virtuals\_vc](https://x.com/@virtuals_vc)[@ethermage](https://x.com/@ethermage)[@Defi0xJeff](https://x.com/@Defi0xJeff)[@flock\_io](https://x.com/@flock_io)[@ronbodkin](https://x.com/@ronbodkin)[@cookiedotfun](https://x.com/@cookiedotfun)[@diego\_defaiPost](https://x.com/@diego_defaiPost)a thread on my account: Start the first tweet of the thread with an overall summary of all content Follow with one tweet per account summarizing their content in Korean Each tweet must be at least 3 sentences, focusing on Base, AI, or both
Output:
{% embed url="<https://x.com/AI_Agent_Korea/status/1949995764643725781>" fullWidth="false" %}
With a single task input, AWESOME\_X\_Agent crawls the data, runs GPT analysis, translates content, and posts a clean, structured thread — entirely automated.
***
## Quick Guide: How to Get Started
To use AWESOME\_X\_Agent, you’ll need to connect it to your X account via API access. Follow these steps:
### 1. Getting X Authentication Details
* Visit:<https://docs.x.com/x-api/getting-started/getting-access>
* Apply for the Free or Basic developer tier
* Configure authentication settings
**Step 1**
<figure><img src="https://csgvlxe6fyz7.sg.larksuite.com/space/api/box/stream/download/asynccode/?code=ODdkNzNhZjg2MjY3OTNiZjBiMDAyMmY1OWYyZDRlNzlfVFdqRzB3R21DWnV1WmE0S0J5RER6ZTF3VWxlcUlDd2ZfVG9rZW46SkhJdmJldHdnb2ptMkN4Rmg5Y2wzMTF2Z3BoXzE3NTQ3MzYxNDg6MTc1NDczOTc0OF9WNA" alt=""><figcaption></figcaption></figure>
**Step 2**
<figure><img src="https://csgvlxe6fyz7.sg.larksuite.com/space/api/box/stream/download/asynccode/?code=YzgzYjhhZjY3NWU3NWI3OWJkMGFjNmQyODRiMTUzYTBfaHhwOUhaQ2kwVzNINHBMU1BLNkRWS3pYeVc5dFdEam9fVG9rZW46QzJaQWJ3OUJnbzNvNzZ4SWFPeWwyQzUwZ29mXzE3NTQ3MzYxNjg6MTc1NDczOTc2OF9WNA" alt=""><figcaption></figcaption></figure>
**Step 3**
<figure><img src="https://csgvlxe6fyz7.sg.larksuite.com/space/api/box/stream/download/asynccode/?code=OTRiZTliM2Q4OWU5ZWFiNjk1YzIxZWMzZDU1ZTkwZTNfOFJLVm00VGhDU0NwYVR1VWNXYWZsUVJrSUc2NXd6UElfVG9rZW46QmdoOWJ6d1h2b0g4dWt4TVJ5bGxMbjU5Z0VlXzE3NTQ3MzYxODU6MTc1NDczOTc4NV9WNA" alt=""><figcaption></figcaption></figure>
**Step 4**
<figure><img src="https://csgvlxe6fyz7.sg.larksuite.com/space/api/box/stream/download/asynccode/?code=NjEyNGZmMWIzNzJiMWFiNDQ5YzViMzE1YjQxNjA3MTlfTnhnZGVMQ25EWllvejlWRERlVzBHZERXSGJHWGtkblJfVG9rZW46UkZzZmJ2SUpIbzBFdkJ4NGhFbmxvYkU5Z3dlXzE3NTQ3MzYxOTY6MTc1NDczOTc5Nl9WNA" alt=""><figcaption></figcaption></figure>
### 2. Run AWESOME\_X\_Agent
* Visit:<https://mcp.awenetwork.ai/agent/public-agents>
* Select **AWESOME\_X\_Agent**<br>
<figure><img src="https://csgvlxe6fyz7.sg.larksuite.com/space/api/box/stream/download/asynccode/?code=OGM3OWZhOWIzYmQwYjBhNTk5YTdmOWZhNzc0NWIyYjRfTnFvZjI1YUhRNUQ3NVhoczZ2MGxkVnN3bmtvejNWZVlfVG9rZW46R0lKTGJ4aGl6b01Kc3B4SmRPWmwwUVJ6Z01nXzE3NTQ3MzYyMzQ6MTc1NDczOTgzNF9WNA" alt=""><figcaption></figcaption></figure>
* Enter X auth details into the agent’s config panel, and save auth
<figure><img src="https://csgvlxe6fyz7.sg.larksuite.com/space/api/box/stream/download/asynccode/?code=NzdkMWM5YTVhZjgzZjI4NDIzNzRhM2I0NjE5YjZlNzVfdEY3WUZoOVE4eHBoNktSQ2tjQXE4dEoyZkprc1pjMFNfVG9rZW46V2pJWmJzVnd3b0w1SG94MnhQZGx1MHZpZ0JkXzE3NTQ3MzYyMzQ6MTc1NDczOTgzNF9WNA" alt=""><figcaption></figcaption></figure>
* Enter your task prompt<br>
<figure><img src="https://csgvlxe6fyz7.sg.larksuite.com/space/api/box/stream/download/asynccode/?code=Mzc3NzVlZGIzYzhhNjYyMmQ3ZTAxMjQ0NWY0YmE0MjVfdnN0SVR6STFmb0oxajljNm53bTZ4WmRiQng4YWxWUG9fVG9rZW46SjB4QWJVYkVxb2RxSjF4akZ5Y2x3cmlIZ0hjXzE3NTQ3MzYyMzQ6MTc1NDczOTgzNF9WNA" alt=""><figcaption></figcaption></figure>
You’re now ready to run, analyze, and post.
## Special Notes
Before using AWESOME\_X\_Agent, make sure to:
* Disable 2FA on your X account for the APIs to authenticate correctly
* Be aware of rate limits under X’s free tier: Maximum 17 tweets per 24 hours (including replies and threads)
We recommend using a dedicated account for automation to avoid limits on your main profile. AWESOME\_X\_Agent is now live on the [AWESOME Agent Marketplace](https://mcp.awenetwork.ai/). Run it, or extend it into your own branded content automation bot today.
# API Service
## External API Documentation (v1.0)
Welcome to our External API. This API lets you programmatically interact with our Agent and Task systems to integrate powerful automation and intelligence into your own applications.
### Table of Contents
* [Basics](#basics)
* [API Base Path](#api-base-path)
* [API Version](#api-version)
* [Authentication](#authentication)
* [Common Endpoints](#common-endpoints)
* [Health Check](#health-check)
* [API Info](#api-info)
* [Agent API](#agent-api)
* [List Agents](#list-agents)
* [Get Agent Details](#get-agent-details)
* [Execute Agent (Non-Streaming)](#execute-agent-non-streaming)
* [Execute Agent (Streaming)](#execute-agent-streaming)
* [Cancel Agent Execution](#cancel-agent-execution)
* [Chat with Agent](#chat-with-agent)
* [Create Agent from Task](#create-agent-from-task)
* [Task API](#task-api)
* [List Tasks](#list-tasks)
* [Get Task Details](#get-task-details)
* [Create Task](#create-task)
* [Analyze Task](#analyze-task)
* [Execute Task (Non-Streaming)](#execute-task-non-streaming)
* [Execute Task (Streaming)](#execute-task-streaming)
* [Cancel Task Execution](#cancel-task-execution)
* [Get Task Execution History](#get-task-execution-history)
* [Error Handling](#error-handling)
* [Rate Limiting](#rate-limiting)
* [Usage Examples](#usage-examples)
***
### Basics
#### API Base Path
All API requests are prefixed with:
```plaintext
/api/external
```
#### API Version
Current API version is `1.0`. All responses include header `X-API-Version: 1.0`.
***
### Authentication
All requests to the external API require an API key.
1. **Obtain an API Key**: Log into our web app, go to your user settings, and create a new API key under API Key Management.
2. **Use the API Key**: Include header `X-API-Key` with your key in every request.
**Example Request Headers**:
```http
GET /api/external/agents HTTP/1.1
Host: your-domain.com
Content-Type: application/json
X-API-Key: YOUR_API_KEY_HERE
```
***
### Common Endpoints
#### Health Check
Check the API service status.
* **URL**: `/health`
* **Method**: `GET`
* **Success (200 OK)**:
```json
{
"success": true,
"data": {
"status": "healthy",
"timestamp": "2023-10-27T10:00:00.000Z",
"version": "1.0",
"service": "mcp-server-external-api"
}
}
```
#### API Info
Get basic information about the API.
* **URL**: `/info`
* **Method**: `GET`
* **Success (200 OK)**:
```json
{
"success": true,
"data": {
"name": "MCP Server External API",
"version": "1.0",
"description": "External API for MCP Server Agent execution and management",
"endpoints": {
"agents": "/api/external/agents",
"tasks": "/api/external/tasks"
}
}
}
```
***
### Agent API
Agent API provides a high-level interface to execute pre-configured automation workflows.
#### List Agents
* **URL**: `/agents`
* **Method**: `GET`
* **Query Params**:
* `page` (optional): Page number, default `1`.
* `limit` (optional): Items per page, default `20`, max `100`.
* `category` (optional): Filter by category.
* **Success (200 OK)**:
```json
{
"success": true,
"data": {
"agents": [ /* List of Agent objects */ ],
"pagination": {
"page": 1,
"limit": 20,
"total": 5,
"totalPages": 1
}
}
}
```
#### Get Agent Details
* **URL**: `/agents/:agentId`
* **Method**: `GET`
* **Success (200 OK)**:
```json
{
"success": true,
"data": {
"agent": { /* Agent details */ }
}
}
```
#### Execute Agent (Non-Streaming)
Execute an Agent and return the result when completed.
* **URL**: `/agents/:agentId/execute`
* **Method**: `POST`
* **Request Body**:
```json
{
"input": "I need a market analysis report on the latest AI trends.",
"parameters": { "format": "pdf", "lang": "en" },
"timeout": 180
}
```
* `input` (string, required): The main instruction for the task.
* `parameters` (object, optional): Extra parameters passed to the Agent.
* `timeout` (number, optional): Timeout in seconds, default `120`, max `300`.
* **Success (200 OK)**:
```json
{
"success": true,
"data": {
"taskId": "...",
"agentId": "...",
"result": { /* Execution result */ },
"status": "completed"
}
}
```
#### Execute Agent (Streaming)
Execute an Agent and receive real-time output via Server-Sent Events (SSE).
* **URL**: `/agents/:agentId/execute/stream`
* **Method**: `POST`
* **Request Body**: Same as non-streaming.
* **Response**: `text/event-stream`
* **SSE Event Flow**:
* **`connection_established`**: Connection confirmed.
* ```json
{
"event": "connection_established",
"data": {
"executionId": "external-agent-exec-xxxx",
"agentId": "...",
"agentName": "...",
"input": "...",
"status": "connected"
}
}
```
* **`task_created`**: Internal task created for the Agent execution.
* ```json
{
"event": "task_created",
"data": {
"taskId": "...",
"title": "Agent execution: ..."
}
}
```
* **General streaming data**: Various events forwarded by the Agent intelligent engine (`agentIntelligentService`) reflecting real-time progress.
* **`execution_complete`**: Execution finished.
* ```json
{
"event": "execution_complete",
"data": {
"taskId": "...",
"agentId": "...",
"success": true,
"executionTime": 5432,
"result": { /* Final result */ },
"status": "completed"
}
}
```
* **`cancelled`**: Emitted if the execution is cancelled.
* **`error`**: Emitted if an error occurs.
#### Cancel Agent Execution
* **URL**: `/agents/:agentId/cancel/:executionId`
* **Method**: `POST`
* **Note**: `executionId` comes from the `connection_established` event when streaming.
* **Success (200 OK)**:
```json
{
"success": true,
"data": {
"executionId": "...",
"cancelled": true
}
}
```
#### Chat with Agent
* **URL**: `/agents/:agentId/chat`
* **Method**: `POST`
* **Request Body**:
```json
{
"message": "Hello, what can you do?",
"conversationId": "..."
}
```
* `conversationId` (string, optional): Continue within an existing conversation if provided.
* **Success (200 OK)**:
```json
{
"success": true,
"data": {
"conversationId": "...",
"response": { /* Agent response */ }
}
}
```
#### Create Agent from Task
* **URL**: `/agents`
* **Method**: `POST`
* **Requ프로젝트로 돌아가기
