> For the complete documentation index, see [llms.txt](https://bitget-ai.gitbook.io/hackathon/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://bitget-ai.gitbook.io/hackathon/base-camp-hackathon-s1-en.md).

# Base Camp \* Hackathon S1 EN

> In 2026, the first generation of truly autonomous trading Agents is being written. It's your turn now.
>
> — Bitget AI Team

## 1. The Event

Today we stand at the next threshold: **Agentic Trading**. 

An AI Agent is the first entity that can read news, sentiment, on-chain signals, and macro data — and act autonomously without predefined rules.  A complete loop, no human in the middle. Traditional quant can't do this.

**Bitget AI Base Camp Hackathon S1** is the world's first dedicated competition for AI × Crypto Builders, and the starting point of Bitget's long-term developer program.

<table><thead><tr><th width="179.78125">Item</th><th>Details</th></tr></thead><tbody><tr><td>Duration</td><td>May 27 – June 30, 2026</td></tr><tr><td>Format</td><td>Global, online</td></tr><tr><td>Total Prize Pool</td><td><strong>50,000 USDT</strong></td></tr><tr><td>Organizer</td><td>Bitget</td></tr><tr><td>Strategic Sponsor</td><td>Alibaba Qwen</td></tr><tr><td>Media &#x26; Ecosystem</td><td>Foresight Ventures / Foresight News</td></tr></tbody></table>

### Prizes

<table><thead><tr><th width="224.68359375">Prize</th><th width="165.15234375">Amount</th><th>Track</th></tr></thead><tbody><tr><td>1st Place</td><td>6,600 USDT</td><td>All tracks, judged together</td></tr><tr><td>2nd Place</td><td>1,500 USDT * 3</td><td>1 per track</td></tr><tr><td>3rd Place</td><td>800 USDT * 3</td><td>1 per track</td></tr><tr><td>Community Impact Award</td><td>500 USDT * 3</td><td>1 per track</td></tr><tr><td>Participation Award</td><td>+50 USDT / team</td><td>（Qualifying teams）Submit Demo + qualifying post</td></tr></tbody></table>

**Beyond cash prizes: official channel exposure, ecosystem media coverage, potential incubation and investor introductions, early access to Bitget AI products, and membership in the Bitget AI Builders long-term developer community.**

## 2. How to Participate

### Timeline (UTC+8)

<table><thead><tr><th width="154.01171875">Date</th><th>Milestone</th></tr></thead><tbody><tr><td>May 27, 00:00</td><td>Registration opens, development starts</td></tr><tr><td>Jun 14, 24:00</td><td>Registration closes (first 500 teams get Qwen token credits)</td></tr><tr><td>Jun 15, 00:00</td><td>Submission window opens</td></tr><tr><td>Jun 25, 24:00</td><td>Submission deadline</td></tr><tr><td>Jun 25–29</td><td>Judging</td></tr><tr><td>Jun 30</td><td>Awards — winners notified by email</td></tr></tbody></table>

### What You Need to Do

{% stepper %}
{% step %}
***Register (May 27, UTC+8)***

* Fill out the form on the [official registration page](https://www.bitget.com/zh-CN/campaigns/d8a2a61fd63c4bc2a3c8198ec923da9a) using your Bitget account (linked email required). Solo or team entries welcome; teams up to 5 members.
* Join the [official Telegram community](https://t.me/+o1tYqQ_lXxllYjgy) after registering. First 500 qualifying teams receive Qwen token credits within 24h; join the community to claim MuleRun Token credits (see Chapter 4).
  {% endstep %}

{% step %}
***Pick a track and start building (immediately)***

* Read Chapter 3 to pick your direction, then connect to Bitget Agent Hub or Bitget Playbook and start building (see Chapter 4).
* Demo must be real and runnable — no concept-only presentations. No real capital required; sim trading and backtest records are accepted.
  {% endstep %}

{% step %}
***Post (throughout development)***

* Repost the Bitget [official interaction post](https://x.com/Bitget_AI/status/2062506424085917944?s=20) and publish your own content, tagged **#BitgetHackathon** and @ the Bitget AI official account.
  {% endstep %}

{% step %}
***Submit (Jun 15, 00:00 – Jun 25, 24:00, UTC+8)***

* Bitget shares the submission link via official social media, community, and registration email.
* Fill in the track form and upload your materials; include links to your repost of the [official interaction post](https://x.com/Bitget_AI/status/2062506424085917944?s=20) to qualify for the Community Impact Award. Submission requirements by track are in Chapter 3.
  {% endstep %}

{% step %}
***Results (Jun 30)***

* Judges score independently after submission closes. Winners notified by registration email; results published publicly.
  {% endstep %}
  {% endstepper %}

## 3. Tracks, Submission & Judging

### Track Overview

<table><thead><tr><th width="126.01171875">Track</th><th width="205.8359375">Come here if…</th><th>What you can build</th></tr></thead><tbody><tr><td>🟦 <strong>Trading Agent</strong></td><td>You want to run strategies automatically or build tools for Vibe Trading</td><td><ul><li>Natural language-driven contract trading Agent</li><li>BTC adaptive trend + mean-reversion strategy</li><li>Meme coin on-chain signal copy bot</li></ul></td></tr><tr><td>🟩 <strong>Trading Infra</strong></td><td>You want to build better infrastructure or data products for Agents</td><td><ul><li>Paper Trading sandbox</li><li>Agent monitoring dashboard + risk scoring<br>On-chain data dashboard</li><li>Natural language strategy compiler</li></ul></td></tr><tr><td>🟧 <strong>Stock AI Trading</strong></td><td>Want to use AI to trade tokenized US stocks, or solve a real problem around US stock trading</td><td><ul><li>Macro/sentiment Agent for tokenized US assets</li><li>Backtest and deploy strategies on US stock historical data</li><li>Agent that reads Fed signals and auto-rebalances positions</li></ul></td></tr></tbody></table>

### Submission Requirements <a href="#submission-requirements-english" id="submission-requirements-english"></a>

> Submission window: **June 15, 2026 0:00 – June 25, 2026 24:00 (UTC+8)**\
> 📢 **Submission link:** [**https://forms.gle/CEGB6fRtuobD3bCj8**](https://forms.gle/CEGB6fRtuobD3bCj8)

***

#### General Rules (All Tracks) <a href="#tong-yong-gui-ze-suo-you-sai-dao" id="tong-yong-gui-ze-suo-you-sai-dao"></a>

All participants must meet the following baseline requirements before submitting:

* **UID must match registration**: The Bitget UID you submit must match the one used at registration — used for eligibility verification
* **All links must be publicly accessible**: No login required to view any submitted link, including Demo pages, GitHub repos, MuleRun publish pages, GetAgent Studio strategy links, etc. If your Demo requires a login, you must also submit a demo video
* **Thesis must be clearly stated**: Your project description must explain the core logic behind your strategy or tool — a feature list or pure concept pitch is not accepted
* **At least one verifiable usage record**: See per-track checklists below. Required for all tracks except Open Innovation

> ⚠️ **Baseline threshold**: Submissions that fail to meet any of the above will be disqualified and will not enter judging.

***

#### Per-Track Submission Checklist <a href="#ge-sai-dao-ti-jiao-checklist" id="ge-sai-dao-ti-jiao-checklist"></a>

#### 🟦 Track 1 · Trading Agent <a href="#sai-dao-yi-jiao-yi-agent" id="sai-dao-yi-jiao-yi-agent"></a>

Build an AI Agent that autonomously perceives market conditions, makes decisions, executes trades, and manages risk in the crypto market. Any instrument is valid — perpetuals, spot, on-chain protocols, prediction markets, etc.

<table><thead><tr><th width="152.3046875">Material</th><th width="160.23828125">Requirement</th><th>Notes</th></tr></thead><tbody><tr><td>Project description</td><td>Required</td><td>Follow the four-part structure above</td></tr><tr><td>GitHub repo or MuleRun / GetAgent Studio publish link</td><td>Required</td><td>GitHub must be public with a complete README; MuleRun / Studio links must be accessible without login</td></tr><tr><td>Live trading record or paper trading log</td><td>Required (preferred)</td><td>Must include: timestamp, trading pair, direction, price, quantity, account balance change; recommended to upload to GitHub and submit the link</td></tr><tr><td>Backtest report</td><td>Optional, supplementary</td><td>Must include the code or notebook used to generate it; screenshots alone are not accepted; judges reserve the right to reproduce results</td></tr><tr><td>Demo video</td><td>Optional; required if Demo needs login</td><td>Recommended: post to Twitter as a public tweet and submit the link; YouTube and other public video links also accepted; max 3 minutes</td></tr></tbody></table>

***

#### 🟩 Track 2 · Trading Infra <a href="#sai-dao-er-jiao-yi-infra" id="sai-dao-er-jiao-yi-infra"></a>

Build infrastructure that helps Agents perform better or makes trading more efficient for traders. Scope includes but is not limited to: tools or frameworks for Agents, products for traders (monitoring dashboards, visualization tools, etc.), and strategy evaluation and benchmarking systems.

<table><thead><tr><th width="164.15234375">Material</th><th width="139.44140625">Requirement</th><th>Notes</th></tr></thead><tbody><tr><td>Project description</td><td>Required</td><td>Follow the four-part structure above</td></tr><tr><td>GitHub repo</td><td>Required</td><td>Must be public; README must include: installation steps, integration instructions, usage examples; another developer should be able to run it independently using the README</td></tr><tr><td>Deployment link</td><td>Optional</td><td>If an online version is available, include the deployment URL</td></tr><tr><td>Verifiable usage record</td><td>Required</td><td>Submit any one of the following: API call log (with timestamps and call volume) / real user usage records / sample input + output files (reproducible by judges) / description or link showing another developer has integrated it</td></tr><tr><td>Demo video</td><td>Optional; required if Demo needs login</td><td>Recommended: post to Twitter as a public tweet and submit the link; YouTube and other public video links also accepted; max 3 minutes; recommended to show the full flow from installation to core functionality</td></tr></tbody></table>

***

#### 🟧 Track 3 · US Stock AI Trading (US stock projects) <a href="#sai-dao-san-mei-gu-ai-jiao-yi-mei-gu-xiang-guan-xiang-mu" id="sai-dao-san-mei-gu-ai-jiao-yi-mei-gu-xiang-guan-xiang-mu"></a>

Use AI to participate in US stock asset trading, or solve a real problem around the US stock trading experience. Eligible assets include tokenized US stocks on the Bitget platform, as well as traditional US stocks and ETFs.

<table><thead><tr><th width="136.88671875">Material</th><th width="158.046875">Requirement</th><th>Notes</th></tr></thead><tbody><tr><td>Project description</td><td>Required</td><td>Follow the four-part structure above</td></tr><tr><td>GitHub repo or Demo link</td><td>Required</td><td>GitHub must be public with a README; Demo must be accessible without login</td></tr><tr><td>Live trading record or paper trading log</td><td>Required (preferred)</td><td>Must include: timestamp, asset, direction, price, quantity, account balance change; recommended to upload to GitHub and submit the link</td></tr><tr><td>Backtest report</td><td>Optional, supplementary</td><td>Must include the code or notebook used to generate it; screenshots alone are not accepted</td></tr><tr><td>Demo video</td><td>Optional; required if Demo needs login</td><td>Recommended: post to Twitter as a public tweet and submit the link; YouTube and other public video links also accepted; max 3 minutes</td></tr></tbody></table>

***

#### 🟧 Track 3 · US Stock AI Trading (Open Innovation) <a href="#sai-dao-san-mei-gu-ai-jiao-yi-kai-fang-chuang-xin-xiang-mu" id="sai-dao-san-mei-gu-ai-jiao-yi-kai-fang-chuang-xin-xiang-mu"></a>

Participants who registered early under the "Open Innovation" track may submit projects unrelated to US stocks, but the project must have a clear connection to crypto trading or AI, and must use the UID from registration.

<table><thead><tr><th width="116.09765625">Material</th><th width="159">Requirement</th><th>Notes</th></tr></thead><tbody><tr><td>Project description</td><td>Required</td><td>Follow the four-part structure above; must describe the connection to crypto trading or AI</td></tr><tr><td>GitHub repo or Demo link</td><td>Required</td><td>GitHub must be public with a README; Demo must be accessible without login</td></tr><tr><td>Evidence of the project running</td><td>Optional; preferred if available</td><td>Any one of the following: Agent output screenshots + raw files / sample input/output / API call log / demo run record / test data files; recommended to upload to GitHub and submit the link</td></tr><tr><td>Demo video</td><td>Optional; required if Demo needs login</td><td>Recommended: post to Twitter as a public tweet and submit the link; YouTube and other public video links also accepted; max 3 minutes</td></tr></tbody></table>

***

#### Community Impact Award (All Tracks, Optional) <a href="#chuan-bo-jiang-suo-you-sai-dao-xuan-tian" id="chuan-bo-jiang-suo-you-sai-dao-xuan-tian"></a>

Scored independently from the main judging criteria — has no effect on your technical score.

<table><thead><tr><th width="139.3359375">奖项</th><th width="121.46875">材料</th><th>说明</th></tr></thead><tbody><tr><td>Participation Award</td><td>Engagement tweet link</td><td>Submit a working Demo + a qualifying post (quotethe <a href="https://x.com/Bitget_AI/status/2062506424085917944?s=20">Bitget tweet</a> and share  dev log) to be eligible for the Participation Award:   +50 USDT  </td></tr><tr><td>Community Impact Award</td><td>Engagement tweet link</td><td>Tag #BitgetHackathon and @Bitget_AI, and quote the <a href="https://x.com/Bitget_AI/status/2062506424085917944?s=20">Bitget tweet</a> — top entries win the Community Impact Award (500 USDT × 3)</td></tr></tbody></table>

### Scoring Dimensions <a href="#scoring-dimensions" id="scoring-dimensions"></a>

> The projects that impress judges are the ones that genuinely solve a problem and actually run

Judges evaluate holistically across the following dimensions — no fixed weighting formula. **Bitget reserves the right of final interpretation.**

<table><thead><tr><th width="161.0703125">Dimension</th><th>What judges look for</th></tr></thead><tbody><tr><td><strong>Depth of thesis</strong></td><td>Is the core assumption behind the strategy or tool design sound? Has the builder truly thought through this direction?</td></tr><tr><td><strong>Runnability</strong></td><td>Is the Demo genuinely accessible and runnable? Live trading / paper trading > backtest > pure concept</td></tr><tr><td><strong>Completeness</strong></td><td>Does the MVP work end-to-end? Does the solution match the problem? Honest self-assessment is valued over exaggeration</td></tr><tr><td><strong>Novelty &#x26; potential</strong></td><td>Does it do something only AI Agents can do? Is there room for the idea to grow?</td></tr></tbody></table>

### FAQ <a href="#chang-jian-wen-ti" id="chang-jian-wen-ti"></a>

<details>

<summary><strong>Q: I don't have a deployment environment — is GitHub alone enough?</strong></summary>

Yes, but your README must clearly explain how to run it locally. Judges should be able to get it running without contacting you.

</details>

<details>

<summary><strong>Q: Can I submit backtest screenshots?</strong></summary>

No. Backtest results must be accompanied by the code or notebook used to generate them. Judges reserve the right to reproduce the results.

</details>

<details>

<summary><strong>Q: What if my Demo requires a login to access?</strong></summary>

You must also submit a demo video (public Twitter link or YouTube) so judges can see the project running.

</details>

<details>

<summary><strong>Q: What if my MuleRun / GetAgent Studio link requires a login?</strong></summary>

Same as above — you must also submit a demo video.

</details>

<details>

<summary><strong>Q: I don't have live or paper trading data — can I submit with only a backtest?</strong></summary>

Yes, but backtest-only submissions are scored lower than projects with actual run records. Backtests must include the code — screenshots alone are not accepted.

</details>

<details>

<summary><strong>Q: Do Open Innovation projects need to submit a usage record?</strong></summary>

It's optional, but preferred. Accepted formats are flexible: Agent output screenshots, sample input/output, API call logs, demo run records, test data files — any of these work.

</details>

<details>

<summary><strong>Q: Can I submit an Open Innovation project that has nothing to do with US stocks?</strong></summary>

Yes, but you must clearly describe its connection to crypto trading or AI, and use the UID from your registration.

</details>

<details>

<summary><strong>Q: Is Part 4 of the project description (your take on AI Trading) required?</strong></summary>

Optional — it won't affect your base score. But if you have genuine hands-on experience and insights, including them helps judges better understand your project.

</details>

## 4. Hackathon Toolkit

### Bitget Playbook (Crypto & Stock AI Quant Copilot)

Bitget Playbook is an AI-driven quantitative strategy platform. Describe a trading idea in plain language, AI generates a runnable strategy, backtest on real historical data, review PnL / max drawdown / Sharpe / win rate, then deploy live for 24/7 automated execution.

**Option 1: Via the Bitget Official Website**

{% stepper %}
{% step %}
***Create an Agent & Sub-Account***

Log in the [Playbook page](https://www.bitget.com/zh-CN/activity/ai-get-agent/playbook?tab=explore) with your registered Bitget UID, Click "Create Agent" and follow the instructions to set up your sub-account and obtain your Playbook API Key
{% endstep %}

{% step %}
***Use a Coding Agent to Build, Backtest & Publish Your Strategy***

Open Claude Code (or any supported coding agent) and send the following prompt in full:

<pre><code>1. Install getagent using https://www.npmjs.com/package/@bitget-ai/getagent-skill
<strong>2. Use getagent to create a strategy playbook about [your idea], then upload, backtest, and publish it
</strong>3. Once backtest succeeds, show me the key metrics in a table

Strategy philosophy:
[Describe your core logic, e.g. adaptive market regime — trend-following when trending, mean reversion when ranging, stay flat when unclear]

playbook key: [your Playbook API Key]
</code></pre>

{% endstep %}

{% step %}
***View Your Published Playbook***

Once published, your strategy will be visible directly on [here](https://www.bitget.com/zh-CN/activity/ai-get-agent/playbook?tab=explore)
{% endstep %}
{% endstepper %}

**Option 2: Via GetAgent Studio**

{% stepper %}
{% step %}
***Get Whitelist Access***

Join the [official Telegram community](https://t.me/+o1tYqQ_lXxllYjgy) and DM the admin with your UID to get whitelist access
{% endstep %}

{% step %}
**Via GetAgent Studio**

Once approved, you can explore the full Playbook experience at [here](https://getagent.studio/)
{% endstep %}
{% endstepper %}

### [Bitget Agent Hub](https://github.com/BitgetLimited/agent_hub) (Crypto & Stocks Agent's Trading Arsenal)

Bitget Agent Hub is a trading tool platform built for AI developers. Three layers, use what you need:

<table><thead><tr><th width="135.53125">Module</th><th width="283.09375">What it is</th><th>How it helps</th></tr></thead><tbody><tr><td><strong>Tools</strong></td><td>58 trading APIs (spot, futures, account, etc.)</td><td>Your Agent can place orders, check positions, and manage assets without writing any API wrappers</td></tr><tr><td><strong>Skill Hub</strong></td><td>5 analyst-grade perception Skills (macro, sentiment, technicals, on-chain, news)</td><td>Give your Agent market awareness in minutes — just configure and go</td></tr><tr><td><strong>MCP Server</strong></td><td>One-line setup for Claude / Cursor / Codex</td><td>Call all Bitget trading capabilities directly from the AI tools you already use</td></tr></tbody></table>

**Quick install:**

```bash
npx bitget-hub upgrade-all --target claude
```

Deploy to specific tools:

```bash
npx bitget-hub install --target codex
npx bitget-hub install --target claude,codex
```

**Configure Bitget API Key:** Log in to bitget.com → Settings → API Management, create a key with Read and/or Trade permissions:

```bash
export BITGET_API_KEY="your-api-key"
export BITGET_SECRET_KEY="your-secret-key"
export BITGET_PASSPHRASE="your-passphrase"
```

**Connect via MCP Server (Claude Code / Cursor):**

```bash
claude mcp add -s user \
  --env BITGET_API_KEY=your-api-key \
  --env BITGET_SECRET_KEY=your-secret-key \
  --env BITGET_PASSPHRASE=your-passphrase \
  bitget \
  -- npx -y bitget-mcp-server
```

**Skill Hub — 5 ready-to-use analyst-grade Skills:**

<table><thead><tr><th width="189.6171875">Skill</th><th>Capability</th></tr></thead><tbody><tr><td><code>macro-analyst</code></td><td>Macro &#x26; cross-asset: Fed policy, BTC vs DXY / Nasdaq / Gold</td></tr><tr><td><code>market-intel</code></td><td>On-chain &#x26; institutional: ETF flows, whale activity, DeFi TVL</td></tr><tr><td><code>news-briefing</code></td><td>News aggregation &#x26; narrative synthesis: briefings, keyword search</td></tr><tr><td><code>sentiment-analyst</code></td><td>Sentiment &#x26; positioning: Fear &#x26; Greed, long/short ratios, funding rates</td></tr><tr><td><code>technical-analysis</code></td><td>Technical analysis: 23 indicators across 6 categories</td></tr></tbody></table>

### Qwen Token Credits

First 500 registered teams receive Alibaba Cloud Qwen token credits.

<table><thead><tr><th width="156.09765625">Item</th><th>Details</th></tr></thead><tbody><tr><td>When</td><td>Within 24h of registration, sent to the email linked to your Bitget UID</td></tr><tr><td>Works with</td><td>Cursor, Codex, and other major coding agents (<strong>Claude Code not supported</strong>)</td></tr><tr><td>Didn't receive it?</td><td>@ an admin in the <a href="https://t.me/+o1tYqQ_lXxllYjgy">official Telegram community</a> with your UID and registration time</td></tr></tbody></table>

> ⚠️ Qwen tokens only work with Alibaba Cloud APIs. They cannot be used with Claude or any non-Alibaba model.

#### Qwen Setup Guide

<table><thead><tr><th width="231.6875">Setting</th><th>Value</th></tr></thead><tbody><tr><td>Base URL</td><td><code>https://hackathon.bitgetops.com/v1</code></td></tr><tr><td>Model (recommended)</td><td><code>qwen3.6-plus</code></td></tr><tr><td>Optional models</td><td><code>qwen3.6-flash</code> (fast)</td></tr></tbody></table>

> ⚠️ Model IDs use hyphens (e.g. `qwen3.6-plus`). Do not use `qwen3.6 max` (with a space).

#### Qwen in Codex

{% stepper %}
{% step %}
***Codex Step 1: Edit config.toml***

In Codex App: Settings → Config → Open config.toml.&#x20;

<div><figure><img src="/files/fhwZz1kardRuCETnCL6N" alt=""><figcaption></figcaption></figure> <figure><img src="/files/MT8yrZiSdc6bpusztiCQ" alt=""><figcaption></figcaption></figure></div>

Add the following at the top (`model_provider` may not exist by default):

```toml
model = "qwen3.6-plus"
model_provider = "bitget-qwen"

[model_providers.bitget-qwen]
name = "Bitget Qwen"
base_url = "https://hackathon.bitgetops.com/v1"
env_key = "BITGET_QWEN_API_KEY"
wire_api = "responses"
```

{% endstep %}

{% step %}
***Codex Step 2: Set the API Key***

```bash
## Do not write the key into config.toml. Run in Terminal:
launchctl setenv BITGET_QWEN_API_KEY 'your real key'

## Verify: output means the key is set.
launchctl getenv BITGET_QWEN_API_KEY

```

<figure><img src="/files/yji4aTcdSIUZdrcmPUAk" alt=""><figcaption></figcaption></figure>
{% endstep %}

{% step %}
***Codex Step 3: Restart Codex App***

Complete Step 2 first, then fully quit Codex (Cmd+Q, not just close the window) and reopen.
{% endstep %}

{% step %}
***Codex Step 4: Confirm you're using Qwen***

* Check the bottom-right corner: should show `Bitget Qwen`, not GPT-5.5
* Terminal: `/Applications/Codex.app/Contents/Resources/codex doctor | grep model` — expected output includes `qwen3.6-plus`

<figure><img src="/files/ocnHVmgqleIB2aCfptMG" alt=""><figcaption></figcaption></figure>
{% endstep %}
{% endstepper %}

#### Qwen in Cursor

{% stepper %}
{% step %}
***Cursor Step 1: Open model settings***

Cursor Settings, or `Cmd+Shift+J` (macOS) / `Ctrl+Shift+J` (Windows), select Models in the left sidebar.

<figure><img src="/files/0hQnztlb6UfLLkScNIgb" alt=""><figcaption></figcaption></figure>
{% endstep %}

{% step %}
***Cursor Step 2: Enter the API Key &*****&#x20;Enable Base URL override**

* Paste your key into the OpenAI API Key field. Click Verify to confirm.
* Turn on Override OpenAI Base URL and enter (The `/v1` suffix is required.)

```
https://hackathon.bitgetops.com/v1
```

<figure><img src="/files/CEEwAoskbeMFx1FkHwO9" alt=""><figcaption></figcaption></figure>
{% endstep %}

{% step %}
***Cursor Step 4: Add custom models***

Click `+ Add model`, enter `qwen3.6-plus`, press Enter and ensure it's enabled. For the fast variant, add `qwen3.6-flash`.

<div><figure><img src="/files/E8kKYYvMwVrrvhOJ1N7A" alt=""><figcaption></figcaption></figure> <figure><img src="/files/MZyEDQ9LjBN2Jwf7FYPI" alt=""><figcaption></figcaption></figure></div>
{% endstep %}

{% step %}
***Cursor Step 5: Select the model and test***

Open Chat / Agent, select `qwen3.6-plus` from the model dropdown, send a test message to confirm it works.

<figure><img src="/files/G4Jv7h4YKJbEoSyFYHVE" alt=""><figcaption></figcaption></figure>
{% endstep %}
{% endstepper %}

### MuleRun Token Credits

Join the official Telegram community to claim 2,000 Credits.

<table><thead><tr><th width="156.2421875">Item</th><th>Details</th></tr></thead><tbody><tr><td>How to claim</td><td>Join the <a href="https://t.me/+o1tYqQ_lXxllYjgy">official Telegram community</a></td></tr><tr><td>Works with</td><td>MuleRun platform, covers major LLMs</td></tr><tr><td>Limit</td><td>One claim per email</td></tr></tbody></table>

**Claim steps:**

{% stepper %}
{% step %}
***Join the official Telegram community***

Join the [official Telegram community](https://t.me/+o1tYqQ_lXxllYjgy)
{% endstep %}

{% step %}
***Sign up at mulerun.com***

Sign up at [mulerun.com](https://mulerun.com/?utm_campaign=0526bitget)
{% endstep %}

{% step %}
***Claim Credits***

Go to [credits.mule.page](https://credits.mule.page/), enter code `0526BITGET` to claim 2,000 Credits
{% endstep %}
{% endstepper %}

> ⚠️ MuleRun Credits must be used on the MuleRun platform. After building on MuleRun, you can submit your Agent link directly as your hackathon Demo.

### Resources

<table><thead><tr><th width="212.484375">Resource</th><th>Link</th></tr></thead><tbody><tr><td>Agent Hub GitHub</td><td><a href="https://github.com/BitgetLimited/agent_hub">BitgetLimited/agent_hub</a></td></tr><tr><td>Event Landing Page</td><td><a href="https://bitget.com/zh-CN/activity-hub/hackathon">bitget.com/zh-CN/activity-hub/hackathon</a></td></tr><tr><td>Registration</td><td><a href="https://www.bitget.com/zh-CN/campaigns/d8a2a61fd63c4bc2a3c8198ec923da9a">Register here</a></td></tr><tr><td>Official Telegram</td><td><a href="https://t.me/+o1tYqQ_lXxllYjgy">Join community</a></td></tr><tr><td>Official interaction post</td><td><a href="https://x.com/Bitget_AI/status/2062506424085917944?s=20">Repost to participate</a></td></tr><tr><td>Submission link</td><td><a href="https://forms.gle/mGrppbHw6gz5jmot8">https://forms.gle/mGrppbHw6gz5jmot8</a></td></tr></tbody></table>

## 5. FAQ

<details>

<summary>Q: Can I participate without trading experience?</summary>

Yes. The [Bitget Agent Hub GitHub repository](https://github.com/BitgetLimited/agent_hub) provides a complete skeleton — just extend it with your own ideas. Bitget Playbook also lets you run a trading strategy end-to-end without writing code. **No real capital required** — sim trading and backtest records are accepted.

</details>

<details>

<summary>Q: Can I participate solo? How does team registration work?</summary>

Absolutely. Solo and team entries are judged equally. Everyone registers as an individual by default — no need to submit teammate information at registration. Final team details will be collected again at the submission stage. Teams can have up to 5 members. Looking for teammates? Post in the official Telegram community.

</details>

<details>

<summary>Q: What if my registration fails?</summary>

Contact an admin directly in the [official Telegram community](https://t.me/+o1tYqQ_lXxllYjgy).

</details>

<details>

<summary>Q: How do I use Bitget Agent Hub?</summary>

Three ways, use what fits:

* **MCP Server**: one command to connect Claude Code / Cursor; then call 58 Bitget trading APIs through natural language
* **Skill Hub**: 5 plug-and-play perception modules (macro / on-chain / sentiment / technicals / news) — configure and go
* **Direct tool use**: clone from GitHub for fully custom setups

Detailed setup in Chapter 4.

</details>

<details>

<summary>Q: How do I use Bitget Playbook?</summary>

Two manual steps first (create sub-account with your registered UID + contact admin via Telegram to get a Playbook API Key), then paste a single prompt into Claude Code — the Agent handles install → strategy creation → backtest → publish automatically. Detailed steps in Chapter 4.

</details>

<details>

<summary>Q: How do I claim MuleRun Credits? What if I didn't receive them?</summary>

Join the official Telegram community, sign up on MuleRun, and enter code `0526BITGET` at [credits.mule.page](https://credits.mule.page/) to claim 2,000 Credits. One claim per email. If you didn't receive them, DM an admin with your UID.

</details>

<details>

<summary>Q: How do I claim Qwen token credits? What if I didn't receive them?</summary>

The first 500 registered teams receive an API Key by email within 24h of registration. It works with Cursor and Codex (**Claude Code not supported**). If you didn't receive it, @ an admin in the official Telegram community with your UID and registration time.

</details>

<details>

<summary>Q: The Qwen API Key doesn't work / can't connect to Qwen. What do I do?</summary>

This key routes through a Bitget proxy — it **cannot connect directly to the official Qwen API**. You must configure the correct Base URL (`https://hackathon.bitgetops.com/v1`) as shown in the Qwen Setup Guide in Chapter 4. Alternatively, use your own Qwen account API Key.

</details>

<details>

<summary>Q: How do I qualify for the Community Impact Award?</summary>

Repost the Bitget [official interaction post](https://x.com/Bitget_AI/status/2062506424085917944?s=20) and publish your own content (project ideas, progress, Showcase, etc.) tagged **#BitgetHackathon** and @ the Bitget AI official account. Submit the post links with your project. Submitting a runnable Demo + qualifying post also earns the Participation Award (+50 USDT per team).

</details>

<details>

<summary>Q: Will connecting Agent Hub with my own API Key trigger auto-trades?</summary>

Currently only read access is supported — it syncs positions to a simulated account. Order execution is not fully implemented yet; no real-account trades will be triggered automatically.

</details>

<details>

<summary>Q: What if I hit a technical issue?</summary>

Join the [official Telegram community](https://t.me/+o1tYqQ_lXxllYjgy) — the technical team is online throughout for support.

</details>

***

Bitget AI Base Camp is a long-term program. Hackathon S1 is the beginning, not the end. The Agent you build today may become part of the foundational infrastructure of the Agentic Trading era.

Come build with us. — Bitget AI Team