How to Configure a Custom Morning Brief with OpenClaw: News, Tasks, Content Drafts, and AI Actions by SMS

Start your day with a fully customized morning briefing tailored to your needs—follow this guide to automate news, tasks, AI content drafts, and action reminders using OpenClaw integrated with WisGate's powerful API.

Overview of the Custom Morning Brief Workflow

Imagine waking up to a single SMS containing everything you need to know: today's relevant news filtered to your interests, your task list prioritized by urgency, AI-generated content drafts ready for review, and intelligent action recommendations based on current data. This isn't science fiction—it's a practical workflow you can build today using OpenClaw and WisGate.

The custom morning brief pipeline combines multiple components into one cohesive system. First, multi-source ingestion pulls news and task data from your preferred feeds and management tools. Next, this raw data flows into WisGate's Claude Opus 4.6 model, which generates grounded AI content drafts using current information. The system then produces AI-driven action suggestions based on the briefing content. Finally, the entire compiled briefing is formatted and delivered via SMS to your phone each morning.

What makes this approach powerful is modularity. Each section—news ingestion, task list, AI content drafts, AI action suggestions, and SMS delivery—operates as an independent component. You can enable or disable any module without rewriting your entire configuration. Need to skip news today but keep tasks and drafts? Toggle it off. Want to test SMS delivery without generating new content? Run just that module. This flexibility means your morning brief evolves with your needs.

The workflow leverages OpenClaw's JSON configuration system to define custom AI providers, allowing you to point directly to WisGate's API endpoint. This integration unlocks access to Claude Opus 4.6's 256,000-token context window and advanced reasoning capabilities, all while maintaining cost efficiency through WisGate's competitive pricing structure.

Setting Up OpenClaw Configuration for Your Morning Brief

OpenClaw stores its configuration in a JSON file located in your home directory. This file controls how OpenClaw connects to AI providers, defines available models, and manages execution parameters. To set up your morning brief, you'll edit this configuration to add WisGate as a custom provider.

Open your terminal and navigate to the configuration file using the nano text editor:

nano ~/.openclaw/openclaw.json

Once the file opens, locate the "models" section. This section defines all available AI providers and their associated models. You'll add a new provider entry here that points to WisGate's API endpoint.

The configuration file uses a hierarchical structure. At the top level, you have "providers", which contains individual provider definitions. Each provider specifies a "baseUrl" (the API endpoint), authentication credentials, and an array of available models. For your morning brief, you're adding a provider called "moonshot" that routes requests to WisGate.

Before making changes, take a moment to understand the existing structure. Look for any existing provider entries to see the pattern. This helps you maintain consistency when adding your new WisGate provider. The JSON syntax is strict—missing commas or brackets will cause errors, so careful editing is essential.

When you're ready to edit, position your cursor in the "models" section where you want to add the new provider. If the section is empty or you're adding to existing providers, ensure proper JSON formatting with commas separating entries.

Editing the Models Section to Use WisGate as a Custom Provider

Copy and paste the following configuration into your models section. This defines a custom provider called "moonshot" that points directly to WisGate's API:

"models": {
  "mode": "merge",
  "providers": {
    "moonshot": {
      "baseUrl": "https://api.wisgate.ai/v1",
      "apiKey": "WISGATE-API-KEY",
      "api": "openai-completions",
      "models": [
        {
          "id": "claude-opus-4-6",
          "name": "Claude Opus 4.6",
          "reasoning": false,
          "input": [
            "text"
          ],
          "cost": {
            "input": 0,
            "output": 0,
            "cacheRead": 0,
            "cacheWrite": 0
          },
          "contextWindow": 256000,
          "maxTokens": 8192
        }
      ]
    }
  }
}

Let's break down what each field does. The "baseUrl" field tells OpenClaw where to send API requests—in this case, https://api.wisgate.ai/v1, which is WisGate's API endpoint. The "apiKey" field holds your authentication credential; replace "WISGATE-API-KEY" with your actual WisGate API key. The "api" field specifies the protocol type; "openai-completions" indicates compatibility with OpenAI-style API calls, which WisGate supports.

Within the "models" array, you define Claude Opus 4.6 with specific parameters. The "id" field is "claude-opus-4-6", which is the exact model identifier WisGate uses. The "contextWindow": 256000 means this model can process up to 256,000 tokens of input context—crucial for your morning brief, which may include lengthy news articles and task lists. The "maxTokens": 8192 limits output to 8,192 tokens per response, ensuring concise, focused briefing content.

The "reasoning": false setting disables extended reasoning mode, keeping response times fast and costs low—ideal for a daily automated briefing. The "cost" fields are set to zero because WisGate handles billing separately; these fields don't affect your actual charges but help OpenClaw track usage internally.

After pasting this configuration, verify the JSON syntax carefully. Check that all braces {} and brackets [] are properly matched, and that commas separate all entries except the last item in each object or array. A single syntax error will prevent OpenClaw from loading the configuration.

Integrating WisGate API Calls for AI Content Draft Generation

Once your OpenClaw configuration points to WisGate, you can make API calls to generate AI content drafts. The integration works by sending your briefing data—news summaries, task lists, and context—to Claude Opus 4.6 through WisGate's endpoint.

When you make an API call through OpenClaw to the "claude-opus-4-6" model, the request routes to https://api.wisgate.ai/v1 using your configured API key. WisGate then processes the request and returns a response from Claude Opus 4.6. This happens transparently; you don't need to manually construct HTTP requests. OpenClaw handles the routing based on your configuration.

For your morning brief, you'll structure API calls to include grounding data—current news articles, task descriptions, and relevant context. Grounding means providing the model with specific, factual information to base its responses on. Instead of generating generic content, Claude Opus 4.6 reads your grounding data and produces drafts that directly reference and build upon it.

For example, if your briefing includes three news articles about industry trends, you'd pass those articles as grounding context. Claude Opus 4.6 then generates a summary that synthesizes all three articles, highlighting connections and implications. This produces far more relevant and accurate content than asking the model to generate news summaries from its training data alone.

The 256,000-token context window is essential here. Your morning brief might include dozens of news items, multiple task lists, and substantial background context. With a smaller context window, you'd hit limits quickly. Claude Opus 4.6's large window lets you include comprehensive briefing data in a single API call, reducing latency and improving coherence.

WisGate's stable 20-second response time for base64-encoded outputs from 0.5k to 4k resolution ensures your morning brief generates consistently. You won't experience unpredictable delays that could disrupt your automated schedule. This reliability is critical for a system that needs to deliver your briefing at a specific time each morning.

To structure your API call, you'll typically include a system prompt that defines the briefing format, followed by user content containing your grounding data. The system prompt might specify: "Generate a concise morning briefing with sections for news summary, task priorities, and recommended actions. Keep each section under 200 words." The user content then includes the actual news articles, tasks, and context. Claude Opus 4.6 processes this and returns a formatted briefing ready for SMS delivery.

Managing Modular Sections Independently to Enable or Disable Briefing Components

The power of a modular morning brief lies in flexibility. You don't need to reconfigure your entire pipeline to adjust individual components. Instead, each section—news ingestion, task management, AI content drafts, AI action suggestions, and SMS delivery—can be toggled independently.

To implement this modularity, structure your OpenClaw configuration with conditional logic for each briefing component. You might add a "briefing" section in your JSON that looks like this:

"briefing": {
  "modules": {
    "news": {
      "enabled": true,
      "sources": ["rss_feed_1", "rss_feed_2"]
    },
    "tasks": {
      "enabled": true,
      "source": "task_manager_api"
    },
    "contentDrafts": {
      "enabled": true,
      "model": "claude-opus-4-6",
      "provider": "moonshot"
    },
    "actionSuggestions": {
      "enabled": true,
      "model": "claude-opus-4-6",
      "provider": "moonshot"
    },
    "smsDelivery": {
      "enabled": true,
      "phoneNumber": "+1-555-0123"
    }
  }
}

With this structure, you can disable the news module by changing "enabled": true to "enabled": false without touching any other configuration. Your morning brief will skip news ingestion but continue with tasks, drafts, actions, and SMS delivery.

This modularity also supports testing. Before deploying your full morning brief, you might enable only the SMS delivery module to verify that messages send correctly. Then enable content drafts to test AI generation. Finally, enable all modules for production use.

To edit these settings, open your configuration file again with nano ~/.openclaw/openclaw.json, locate the "briefing" section, and change the "enabled" values as needed. Each module operates independently, so disabling one doesn't affect others.

You can also adjust module-specific parameters without affecting the overall pipeline. For instance, if you want to add a new news source, modify the "sources" array under the news module. If you want to change which model generates content drafts, update the "model" field under contentDrafts. These changes are isolated and don't require reconfiguring the entire system.

This approach scales well. As your morning brief evolves—perhaps you add weather data, stock prices, or calendar events—you simply add new modules with their own "enabled" flags. The existing modules continue working unchanged.

Deploying and Testing Your SMS Delivery of the Morning Brief

Once your configuration is complete, save your changes and restart OpenClaw to load the new settings. In the nano editor, save by pressing Ctrl + O, then confirm by pressing Enter. Exit the editor by pressing Ctrl + X.

Now restart OpenClaw. First, stop the current process by pressing Ctrl + C in your terminal. Then launch OpenClaw's terminal user interface:

openclaw tui

OpenClaw will reload your configuration, including the new WisGate provider and briefing modules. If there are any JSON syntax errors, OpenClaw will report them in the terminal output. Fix any errors by reopening the configuration file and correcting the problematic lines.

Once OpenClaw starts successfully, test your SMS delivery. Trigger a manual briefing execution to verify that the system generates content and sends an SMS. Monitor the terminal output for any errors during execution. Check your phone to confirm the SMS arrives with the expected content.

During testing, you might encounter issues. If SMS delivery fails, verify that your phone number is correctly configured and that your SMS provider credentials are valid. If content generation seems slow, check that your WisGate API key is active and that your internet connection is stable. If the briefing content looks incomplete, review your grounding data to ensure all necessary information is included in the API call.

Test each module independently before running the full pipeline. Disable all modules except SMS delivery and trigger a test run. This isolates SMS functionality from content generation issues. Then enable content drafts and test again. This methodical approach helps you identify and fix problems quickly.

Once all modules pass individual testing, enable the full pipeline and run a complete morning brief test. Verify that news, tasks, drafts, actions, and SMS all work together correctly. If everything looks good, schedule your morning brief to run automatically each morning at your preferred time.

Pricing and Performance Benefits with WisGate API

WisGate offers significant cost advantages compared to official API pricing. For image generation, WisGate charges $0.058 USD per image, while the official rate is $0.068 USD per image. This 15% savings compounds quickly if your morning brief includes generated images—product mockups, data visualizations, or illustrated summaries.

Beyond pricing, WisGate delivers consistent performance. Image generation from 0.5k to 4k resolution in base64 format completes in approximately 20 seconds. This predictable response time is crucial for automated systems. You can reliably schedule your morning brief to complete within a specific window, ensuring it arrives at your preferred time each morning.

Claude Opus 4.6 through WisGate provides the performance characteristics your briefing needs. The 256,000-token context window accommodates extensive briefing data without requiring multiple API calls. The 8,192 maximum tokens per response ensures output stays concise and focused. The model's reasoning capabilities, when enabled, can analyze complex briefing data and generate nuanced action recommendations.

For a morning brief that runs daily, the cost difference between WisGate and official pricing adds up. Over a year, the 15% savings on image generation alone could amount to hundreds of dollars, depending on your usage. More importantly, WisGate's stable performance means your briefing arrives reliably each morning without unexpected delays.

WisGate's infrastructure is optimized for reliability. The API endpoint https://api.wisgate.ai/v1 is designed to handle high-volume requests without degradation. For your morning brief, this means consistent performance whether you're running it alone or alongside thousands of other users.

The combination of cost efficiency and performance makes WisGate an ideal choice for automated briefing systems. You get enterprise-grade AI capabilities at a price point that makes daily automation economically sensible.

Summary and Next Steps

You now have a complete roadmap for building a custom morning brief with OpenClaw and WisGate. The process involves configuring OpenClaw to recognize WisGate as a custom AI provider, structuring your briefing pipeline with modular components, and deploying the system with SMS delivery.

The key steps are straightforward: edit ~/.openclaw/openclaw.json to add the WisGate provider with Claude Opus 4.6, structure your briefing modules with independent enable/disable flags, test each component individually, and deploy the full pipeline. The modularity ensures you can adjust your briefing as your needs evolve without rebuilding the entire system.

Begin implementing your morning brief now. Start by exploring WisGate's capabilities at https://wisgate.ai/ and https://wisgate.ai/models to understand the full range of available models and pricing options. Then follow the configuration steps provided here to integrate Claude Opus 4.6 in OpenClaw. Test your setup thoroughly before scheduling automated runs. Once everything works reliably, you'll have a powerful system delivering personalized briefings to your phone each morning.