RFD Hot Deals Discord Bot

The ultimate companion for your Discord server to catch the hottest deals from RedFlagDeals!

Adding this bot to your server will automatically post the latest and greatest deals directly into the channel of your choice. Never miss another price error or "Lava Hot" deal again.

Tip

Want to use the bot right away? Click the link below to invite the bot to your server!

👉 Invite the RFD Bot to Your Discord Server

(The bot only requests permission to Send Messages and Embed Links)

🚀 How to Setup and Use

Invite the bot: Click the invite link above and select the server you want to add it to.
Choose a channel: Go to the channel where you want the bot to post deals.
Run the setup command: Type /rfd-bot-setup set channel:#your-channel-name type:your-preference and hit Enter!
- The type option allows you to choose exactly what deals you want:
  - All deals
  - Tech only deals
  - All warm + hot deals
  - All warm + hot tech deals
  - All hot deals
  - All hot tech deals
- Note: "Warm" deals require at least 2 likes and high organic engagement. "Hot" deals are determined entirely by the Gemini AI's "Lava Hot" analysis.

That's it! The bot will now monitor the RedFlagDeals "Hot Deals" forum 24/7 and post beautifully formatted alerts directly to that channel based on your chosen type.

To stop the bot from posting or securely view active subscriptions, an administrator can simply type /rfd-bot-setup remove in any channel. This brings up an interactive message with delete buttons for each active channel. Clicking a button will instantly remove that channel's subscription and update the message to confirm the removal, ensuring a clean and responsive UX.

🌟 Features

Identifies RFD Hot Deals: Scrapes the latest deals from the Hot Deals forum natively with fallback support.
Gemini AI Integration: Uses Google Gemini AI (gemini-2.5-flash-lite) with Google Search Grounding to:
- Clean up and summarize messy deal titles based on extracted deal details.
- Determine if a deal is "Lava Hot" (adds 🔥 emojis and escalates the Discord alert color to hot pink).
- Note: Implements exponential backoff to handle free-tier API rate limits gracefully.
Deep Scraping: Extracts detailed deal content, including descriptions and comments from deal pages for AI context.
Discord Bot Notifications: Sends detailed notifications to multiple subscribed Discord servers, complete with native Discord timestamps, actual deal URLs, concise engagement metrics, and categorized emojis for improved visual clarity (e.g. 💻, 🛒, 🍔).
Smart Deduplication: Automatically detects when identical deals are posted in multiple forum threads by fuzzy matching titles and target URLs. It gracefully merges their engagement metrics, sorts the threads by popularity, and appends all tracking links into a single unified Discord alert.
Live Updates: Discord embed colors escalate to warm based on deal metrics (likes/views) or to hot based on AI analysis, and will retain the highest "heat" rank achieved even if the ratios drop later. Embeds are dynamically patched to keep likes, comments, and views accurate for up to 1 hour after publication to respect Discord's rate limits on editing old messages.
Zero-Config For Users: Just invite, set the channel via a Slash Command, and enjoy the deals.

🛠️ For Developers & Self-Hosting

Want to host your own instance or modify the code? The bot operates with a simple, serverless architecture on Google Cloud:

The bot operates with a simple, serverless architecture on Google Cloud:

Cloud Scheduler: Triggers the bot every minute via an HTTP GET to /.
Cloud Run: Hosts the Go HTTP server. When triggered:
1. Scrapes the RFD Hot Deals list natively.
2. Checks Firestore and compares scraped deals with previously processed ones.
3. Fetches deep detail pages (descriptions, comments) for new or updated deals.
4. (Optional) Passes deep deal context to Gemini AI to generate a clean title and "hotness" rating.
5. Sends or updates formatted deal notifications to all subscribed Discord channels via the Bot Token API.
6. Updates Firestore state for deals and active message IDs.
Discord API (Interactions): Exposes an endpoint (/discord/interactions) to handle Slash Command setups and removals securely (via Ed25519 signature validation).
Firestore: Stores deal records and active server subscriptions (guildID -> channelID).

graph LR
    A[Cloud Scheduler: Runs every minute] --> B(Cloud Run: Go Server)
    B --> C{Scrape RFD List}
    C --> D{Check/Fetch heavily detailed info}
    D --> E{Gemini AI Analysis}
    E --> F{Format & Post to Discord}
    F --> G[Update Firestore State]

Local Development

Clone Repository

git clone https://your-repository-url/rfd-discord-bot.git
cd rfd-discord-bot

(Replace https://your-repository-url/rfd-discord-bot.git with the actual URL of your repository)

Environment Setup (Go)

Install Go: Download and install Go from the official Go download page.
Navigate to Directory: Open your terminal and navigate to the cloned rfd-discord-bot directory.
```
cd path/to/rfd-discord-bot
```
Install Dependencies: Run the following command to download and install the necessary Go modules:
```
go mod tidy
```

Environment Variables for Local Testing

The application requires the following environment variables to run locally:

GOOGLE_CLOUD_PROJECT: Your Google Cloud Project ID.
DISCORD_APP_ID: The Discord Application ID (for Slash Command registration).
DISCORD_PUBLIC_KEY: The Discord Public Key (for Ed25519 Interaction signature validation).
DISCORD_BOT_TOKEN: The Discord Bot Token used for API authorization.
GEMINI_API_KEY: (Optional) Your Google Gemini API key. If omitted, AI features like grounded title cleaning and hotness rating are disabled.
PORT: (Optional) The port the HTTP server should listen on. Defaults to 8080.

It's recommended to create a .env file in the project root to store these variables. This file is included in .gitignore to prevent accidental commits of sensitive information.

Create .env file:

# .env
export GOOGLE_CLOUD_PROJECT="your-gcp-project-id"
export DISCORD_APP_ID="your-discord-app-id"
export DISCORD_PUBLIC_KEY="your-discord-public-key"
export DISCORD_BOT_TOKEN="your-discord-bot-token"
export GEMINI_API_KEY="your-gemini-api-key"
export PORT="8080"

Replace placeholder values with your real tokens from the Discord Developer Portal and Google Cloud.

Load Environment Variables: Before running the application, source the .env file:

source .env

Alternatively, you can set these variables directly in your shell session:

export GOOGLE_CLOUD_PROJECT="your-gcp-project-id"
export DISCORD_APP_ID="your-discord-app-id"
export DISCORD_PUBLIC_KEY="your-discord-public-key"
export DISCORD_BOT_TOKEN="your-discord-bot-token"
export GEMINI_API_KEY="your-gemini-api-key"

Running Locally

Once the environment variables are set, you can run the application using:

go run main.go

Or, if all your Go files are in the main package and in the root directory:

go run .

This will start an HTTP server, typically on port 8080 (or the port specified by the PORT environment variable, which Cloud Run sets automatically).

To test the handler locally, you can send a GET request to http://localhost:8080/ (or the specific path your handler listens on, if different) using a tool like curl or your web browser.

curl http://localhost:8080/

Note: While you can trigger it via HTTP locally, in the deployed Google Cloud environment, the primary trigger is Cloud Scheduler.

Running Tests

To run the standard unit tests:

go test ./...

To run the integration tests (which spin up a local mock server and test the full pipeline):

go test ./internal/processor/... -tags=integration -v

Google Cloud Deployment (Detailed & Hand-holding)

These instructions will guide you through deploying the bot to Google Cloud Run.

Prerequisites

Google Cloud Project: You need an active Google Cloud Project. If you don't have one, create one at the Google Cloud Console.
gcloud CLI: Install and initialize the Google Cloud CLI.
- Installation instructions: Google Cloud SDK
- Authenticate and initialize gcloud:
```
gcloud auth login
gcloud init
```
  During gcloud init, you will be prompted to pick a project and can set a default region. This is the recommended first step.
- Set Environment Variables for gcloud commands: For convenience, and to avoid repeatedly typing your project ID and region, set the following environment variables in your shell. These will be used in subsequent gcloud commands.
  
  For Linux/macOS (bash/zsh):
```
export PROJECT_ID=$(gcloud config get-value project)
export REGION="us-central1" # Or your preferred region, e.g., us-east1, europe-west1
echo "PROJECT_ID set to: $PROJECT_ID"
echo "REGION set to: $REGION (Ensure this is a valid region for your services and that it supports Cloud Run, Firestore, and Cloud Scheduler)"
```
  For Windows (Command Prompt):
```
for /f "tokens=*" %i in ('gcloud config get-value project') do set PROJECT_ID_VAL=%i
set REGION_VAL=us-central1
echo PROJECT_ID_VAL is set to: %PROJECT_ID_VAL%
echo REGION_VAL is set to: %REGION_VAL% (Ensure this is a valid region for your services)
```
  Then use %PROJECT_ID_VAL% and %REGION_VAL% in the gcloud commands where indicated.
  
  For Windows (PowerShell):
```
$env:PROJECT_ID = $(gcloud config get-value project)
$env:REGION = "us-central1" # Or your preferred region
Write-Host "PROJECT_ID set to: $env:PROJECT_ID"
Write-Host "REGION set to: $env:REGION (Ensure this is a valid region for your services)"
```
  Then use $env:PROJECT_ID and $env:REGION in the gcloud commands where indicated.
  
  Note on Region Selection: While us-central1 is suggested as a common, cost-effective default, it's crucial to:
  1. Verify that your chosen region supports all necessary services (Cloud Run, Firestore, Cloud Scheduler).
  2. Consider factors like latency to your users or the source of data (e.g., RFD servers).
  3. Check current pricing for services in that region. You can list available compute regions using gcloud compute regions list. The region for Firestore is chosen during its setup and should ideally match your Cloud Run and Cloud Scheduler region.

Enable APIs

The bot requires several Google Cloud APIs to be enabled for your project:

Cloud Run API: For deploying and running the serverless application.
Cloud Firestore API: For database storage.
Cloud Scheduler API: For triggering the bot periodically.
Cloud Build API: For building container images when deploying from source.

Enable them using the following gcloud command:

# For Linux/macOS (ensure PROJECT_ID is set as shown above):
gcloud services enable run.googleapis.com firestore.googleapis.com cloudscheduler.googleapis.com cloudbuild.googleapis.com --project "$PROJECT_ID"

# For Windows CMD (ensure PROJECT_ID_VAL is set):
# gcloud services enable run.googleapis.com firestore.googleapis.com cloudscheduler.googleapis.com cloudbuild.googleapis.com --project "%PROJECT_ID_VAL%"

# For PowerShell (ensure $env:PROJECT_ID is set):
# gcloud services enable run.googleapis.com firestore.googleapis.com cloudscheduler.googleapis.com cloudbuild.googleapis.com --project "$env:PROJECT_ID"

Ensure you use the correct variable ($PROJECT_ID, %PROJECT_ID_VAL%, or $env:PROJECT_ID) based on your shell and the setup instructions above.

Configure Cloud Build Service Account Permissions

When deploying to Cloud Run from source (using the --source . flag), Google Cloud Build is utilized to build your container image and push it to Artifact Registry. The Cloud Build service account requires specific IAM permissions to perform these actions. If these permissions are missing, the deployment will likely fail with a PERMISSION_DENIED error during the build process.

1. Identify Your Project Number: You'll need your Google Cloud Project Number (distinct from the Project ID) to correctly identify the Cloud Build service account. You can retrieve it using the following command:

# For Linux/macOS:
gcloud projects describe "$PROJECT_ID" --format='value(projectNumber)'

# For Windows CMD:
# gcloud projects describe "%PROJECT_ID_VAL%" --format='value(projectNumber)'

# For PowerShell:
# gcloud projects describe "$env:PROJECT_ID" --format='value(projectNumber)'

Make a note of the outputted project number (e.g., 123456789012), as you'll need it in the next step.

2. Grant roles/run.builder to the Cloud Build Service Account: The Cloud Build service account, which typically has an email address like YOUR_PROJECT_NUMBER@cloudbuild.gserviceaccount.com, needs the roles/run.builder role on your project to successfully build and deploy to Cloud Run. Grant this role using the command below, ensuring you replace YOUR_PROJECT_ID and YOUR_PROJECT_NUMBER (the value obtained in the previous step):

# For Linux/macOS (replace YOUR_PROJECT_NUMBER with the value from the previous step):
gcloud projects add-iam-policy-binding "$PROJECT_ID" --member="serviceAccount:YOUR_PROJECT_NUMBER@cloudbuild.gserviceaccount.com" --role="roles/run.builder"

# For Windows CMD (replace YOUR_PROJECT_NUMBER):
# gcloud projects add-iam-policy-binding "%PROJECT_ID_VAL%" --member="serviceAccount:YOUR_PROJECT_NUMBER@cloudbuild.gserviceaccount.com" --role="roles/run.builder"

# For PowerShell (replace YOUR_PROJECT_NUMBER):
# gcloud projects add-iam-policy-binding "$env:PROJECT_ID" --member="serviceAccount:YOUR_PROJECT_NUMBER@cloudbuild.gserviceaccount.com" --role="roles/run.builder"

This step is crucial for deployments from source to succeed.

Artifact Registry Note: Additionally, when deploying from source for the first time, gcloud run deploy might prompt you to enable the Artifact Registry API (artifactregistry.googleapis.com) and offer to create a default Docker repository (e.g., cloud-run-source-deploy) in your chosen region. You should allow these operations when prompted, as Artifact Registry is used to store the container images built by Cloud Build.

Configure Google Cloud Build (project.toml)

Because the project places its main package inside the cmd/server/ directory, you need to tell Google Cloud Build where to find the source code to build the container.

Create a file named project.toml in the project root with the following content:

[[build.env]]
  name = "GOOGLE_BUILDABLE"
  value = "./cmd/server"

Ensure this file is committed to your repository before attempting to deploy from source or connecting a GitHub repository for continuous deployment.

Create Firestore Instance

To set up Firestore for the bot, follow these steps in the Google Cloud Console:

Navigate to Firestore: Go to the Firestore section of the Google Cloud Console.
Create Database: If you don't have a Firestore database in your project, click "Create Database".
Select Database ID:
- You'll be prompted to name your database (Database ID).
- Recommendation: Use the default ID, which is typically (default).
- Justification: The application code (firestore_client.go) is configured to use the default database and does not specify a custom ID.
Choose Edition:
- Select the "Standard" edition.
- Justification: This edition is cost-effective for applications with minimal data and low-frequency operations, suitable for this bot.
Select Mode:
- Choose "Native Mode".
- Justification: This is required by the Go client library used in the application and aligns with existing project documentation.
Set Up Security Rules:
- You will be asked to configure security rules.
- Recommendation: Start with Restrictive rules (e.g., the default option that denies all client access).
- Justification: The bot accesses Firestore using server-side authentication (IAM roles for the Cloud Run service account), not client-side SDKs, so client access can be denied for better security.
Choose Location Type:
- Select "Region" for the location type.
- Justification: A regional location offers lower latency and cost compared to multi-region, and is suitable when your Cloud Run service is deployed to a specific region.
Select Region:
- Choose a specific region (e.g., us-central1, europe-west1).
- Recommendation: Select the same region where you plan to deploy your Cloud Run service.
- Justification: Co-locating Firestore and your Cloud Run service minimizes network latency and potential egress costs. This choice is permanent.
Finalize Creation: Click "Create Database".

The bot will automatically create the necessary collections (deals and subscriptions) within Firestore.

Configure Environment Variables for Deployment

The DISCORD_APP_ID, DISCORD_PUBLIC_KEY, DISCORD_BOT_TOKEN, and GEMINI_API_KEY are sensitive pieces of information and should be passed to the Cloud Run service as environment variables during deployment.

CRITICAL: DO NOT COMMIT ACTUAL SECRETS TO YOUR CODE REPOSITORY.

You will provide these values in the deployment command below.

Deploy Cloud Run Service

Navigate to your project's root directory (rfd-discord-bot) in your terminal. Run the following command to deploy the service:

# For Linux/macOS (ensure PROJECT_ID and REGION are set):
gcloud run deploy rfd-discord-bot --source . --platform managed --region "$REGION" --allow-unauthenticated --set-env-vars GOOGLE_CLOUD_PROJECT=$PROJECT_ID,DISCORD_APP_ID=YOUR_APP_ID,DISCORD_PUBLIC_KEY=YOUR_PUBLIC_KEY,DISCORD_BOT_TOKEN=YOUR_BOT_TOKEN,GEMINI_API_KEY=YOUR_GEMINI_API_KEY --project "$PROJECT_ID"

Explanation of variables and placeholders:

$REGION: This should be the Google Cloud region you set up earlier (e.g., us-central1). It's used for the --region flag.
$PROJECT_ID: This is your Google Cloud Project ID, set up earlier.
YOUR_APP_ID, YOUR_PUBLIC_KEY, YOUR_BOT_TOKEN: Replace with your actual Discord API strings.
YOUR_GEMINI_API_KEY: Replace with your actual Google Gemini API key. If you omit it, AI processing will be gracefully disabled.

After the deployment is successful, the command will output the Service URL. Copy this URL, as you'll need it for setting up Cloud Scheduler.

Set up Cloud Scheduler

Cloud Scheduler will periodically trigger your Cloud Run service to check for new deals.

You can create a Cloud Scheduler job via the GCP Console or using gcloud.

Using gcloud (Recommended):

# For Linux/macOS (ensure PROJECT_ID and REGION are set, replace YOUR_CLOUD_RUN_SERVICE_URL):
gcloud scheduler jobs create http rfd-bot-trigger --location "$REGION" --schedule "* * * * *" --uri YOUR_CLOUD_RUN_SERVICE_URL --http-method GET --time-zone "America/Toronto" --description "Triggers the RFD Discord Bot every minute" --project "$PROJECT_ID"

# For Windows CMD (ensure PROJECT_ID_VAL and REGION_VAL are set, replace YOUR_CLOUD_RUN_SERVICE_URL):
# gcloud scheduler jobs create http rfd-bot-trigger --location "%REGION_VAL%" --schedule "* * * * *" --uri YOUR_CLOUD_RUN_SERVICE_URL --http-method GET --time-zone "America/Toronto" --description "Triggers the RFD Discord Bot every minute" --project "%PROJECT_ID_VAL%"

# For PowerShell (ensure $env:PROJECT_ID and $env:REGION are set, replace YOUR_CLOUD_RUN_SERVICE_URL):
# gcloud scheduler jobs create http rfd-bot-trigger --location "$env:REGION" --schedule "* * * * *" --uri YOUR_CLOUD_RUN_SERVICE_URL --http-method GET --time-zone "America/Toronto" --description "Triggers the RFD Discord Bot every minute" --project "$env:PROJECT_ID"

Explanation of variables and placeholders:

rfd-bot-trigger: A descriptive name for your Cloud Scheduler job.
--location "$REGION": Specifies the region for the Cloud Scheduler job. This MUST be the same region where your Cloud Run service is deployed.
--schedule "* * * * *": Cron syntax for running the job every minute. Adjust as needed (e.g., /2 * * * * for every two minutes, be polite).
--uri YOUR_CLOUD_RUN_SERVICE_URL: Replace this placeholder with the actual Service URL outputted by the gcloud run deploy command.
--http-method GET: The HTTP method used to invoke your Cloud Run service.
--time-zone "America/Toronto": Sets the timezone for interpreting the schedule. Adjust to your local timezone if preferred.
--project "$PROJECT_ID": Specifies the project for the scheduler job.

Using GCP Console:

Go to Cloud Scheduler in the GCP Console.
Click "Create Job".
Configure the job:
- Name: e.g., rfd-bot-trigger
- Region: Choose a region for the scheduler job.
- Description: (Optional) e.g., "Triggers RFD Discord Bot"
- Frequency: Enter the cron schedule (e.g., */5 * * * * for every 5 minutes).
- Timezone: Select your timezone (e.g., America/Toronto).
- Target type: HTTP
- URL: Paste the Cloud Run Service URL.
- HTTP Method: GET (or POST, depending on your handler; GET is fine here).
- Auth header: Select "None" (because we used --allow-unauthenticated for Cloud Run). If you deployed a private service, you'd configure OIDC or OAuth here.
Click "Create".

Configure IAM Permissions

Proper IAM permissions ensure your Cloud services can interact securely.

1. Cloud Run Service Account to Firestore: The Cloud Run service needs permission to read from and write to Firestore.

Find Service Account: When you deploy a Cloud Run service, it uses a service account. By default, this is the Compute Engine default service account (PROJECT_NUMBER-compute@developer.gserviceaccount.com). You can see the exact service account used in the Cloud Run service details page in the GCP Console, under the "Security" or "Details" tab.
Grant Role: Grant this service account the "Cloud Datastore User" role (which includes Firestore permissions) or the more specific "Firestore User" (roles/firestore.user) role.

# For Linux/macOS (replace YOUR_CLOUD_RUN_SERVICE_ACCOUNT_EMAIL with the actual email):
gcloud projects add-iam-policy-binding "$PROJECT_ID" --member="serviceAccount:YOUR_CLOUD_RUN_SERVICE_ACCOUNT_EMAIL" --role="roles/datastore.user"

# For Windows CMD (replace YOUR_CLOUD_RUN_SERVICE_ACCOUNT_EMAIL):
# gcloud projects add-iam-policy-binding "%PROJECT_ID_VAL%" --member="serviceAccount:YOUR_CLOUD_RUN_SERVICE_ACCOUNT_EMAIL" --role="roles/datastore.user"

# For PowerShell (replace YOUR_CLOUD_RUN_SERVICE_ACCOUNT_EMAIL):
# gcloud projects add-iam-policy-binding "$env:PROJECT_ID" --member="serviceAccount:YOUR_CLOUD_RUN_SERVICE_ACCOUNT_EMAIL" --role="roles/datastore.user"

Replace YOUR_CLOUD_RUN_SERVICE_ACCOUNT_EMAIL with the email of the service account your Cloud Run service uses. This is often PROJECT_NUMBER-compute@developer.gserviceaccount.com by default, or a custom one if you configured it. You can find it in the Cloud Run service details in the GCP Console.

2. Cloud Scheduler Service Account to Invoke Cloud Run:

If you used --allow-unauthenticated when deploying your Cloud Run service, this step is generally not needed for the invocation itself, as the service is public.
However, if you deploy your Cloud Run service as private (by omitting --allow-unauthenticated), the Cloud Scheduler service account needs permission to invoke it.
Cloud Scheduler jobs run under a service account. By default, this is the App Engine default service account (YOUR_PROJECT_ID@appspot.gserviceaccount.com). If you use a custom service account for Scheduler, ensure it has the "Cloud Run Invoker" (roles/run.invoker) role for your Cloud Run service.

For a private service, the command would look like:

# Example for a private service - not strictly needed if --allow-unauthenticated was used
# For Linux/macOS (replace YOUR_SCHEDULER_SERVICE_ACCOUNT_EMAIL):
gcloud run services add-iam-policy-binding rfd-discord-bot --member="serviceAccount:YOUR_SCHEDULER_SERVICE_ACCOUNT_EMAIL" --role="roles/run.invoker" --region="$REGION" --platform=managed --project="$PROJECT_ID"

# For Windows CMD (replace YOUR_SCHEDULER_SERVICE_ACCOUNT_EMAIL):
# gcloud run services add-iam-policy-binding rfd-discord-bot --member="serviceAccount:YOUR_SCHEDULER_SERVICE_ACCOUNT_EMAIL" --role="roles/run.invoker" --region="%REGION_VAL%" --platform=managed --project="%PROJECT_ID_VAL%"

# For PowerShell (replace YOUR_SCHEDULER_SERVICE_ACCOUNT_EMAIL):
# gcloud run services add-iam-policy-binding rfd-discord-bot --member="serviceAccount:YOUR_SCHEDULER_SERVICE_ACCOUNT_EMAIL" --role="roles/run.invoker" --region="$env:REGION" --platform=managed --project="$env:PROJECT_ID"

Replace YOUR_SCHEDULER_SERVICE_ACCOUNT_EMAIL with the service account Cloud Scheduler uses (often YOUR_PROJECT_ID@appspot.gserviceaccount.com by default). This step is only necessary if you deployed Cloud Run as a private service (without --allow-unauthenticated).

Usage

Once deployed and Cloud Scheduler is active (it might take a minute for the first scheduled run after creation), the bot runs automatically. New deals from the RFD Hot Deals feed will be processed and posted to your configured Discord channel.

Error Logging & Alerting

Cloud Logging

All standard output from your Go application when running on Cloud Run is automatically captured by Google Cloud Logging.

Structured Logging (slog): When deployed to Cloud Run, the application automatically switches to structured JSON logging using a custom internal/logger package. This maps Go slog levels (Debug, Info, Notice, Warning, Error, Critical, Alert, Emergency) to their proper Google Cloud Logging severity counterparts. It also supports dynamic log-level filtering via the LOG_LEVEL environment variable (e.g. LOG_LEVEL=DEBUG). This is highly beneficial for both GCP log filtering and the Gemini AI integration, as it logs the full AI input prompt and raw output cleanly inside the JSON payload.

To view logs:

Go to the GCP Console.
Navigate to "Logging" > "Logs Explorer".
You can filter logs by your Cloud Run service name:
- In the query builder, select "Cloud Run Revision" as the resource type.
- Then select your service name (rfd-discord-bot) and revision.
To view AI Decisions: Search for jsonPayload.msg="Completed Gemini AI Deal Analysis". You can expand the log entry to see prompt, clean_title, and is_lava_hot fields nicely formatted.

This is the primary place to check for errors or operational messages from your bot.

Critical Alerts (How to Set Up)

For proactive monitoring, you can set up alerts in Google Cloud Monitoring for critical errors.

Navigate to Alerting: In the GCP Console, go to "Monitoring" > "Alerting".
Create Log-Based Metrics (Optional but Recommended for Specific Errors):
- Go to "Logging" > "Log-based Metrics".
- Click "Create Metric".
- Define a filter for specific error messages you want to track (e.g., resource.type="cloud_run_revision" AND resource.labels.service_name="rfd-discord-bot" AND severity=ERROR AND textPayload:"Failed to fetch feed").
- Give the metric a name (e.g., rfd-bot-fetch-errors).
Create an Alert Policy:
- In "Monitoring" > "Alerting", click "+ Create Policy".
- Add Condition:
  - Select the metric you want to alert on. This could be a general error count for your Cloud Run service or a specific log-based metric you created.
  - For example, to alert on any error logs from your service:
    - Target: Cloud Run Revision > Log Entries
    - Filter for your service and severity=ERROR.
    - Configure the condition (e.g., "Alert if count of error logs > 0 in 5 minutes").
  - If using a log-based metric:
    - Target: Search for your custom log-based metric name.
    - Configure the condition (e.g., "Alert if value of metric > 0 for 5 minutes").
- Notifications:
  - Choose or create notification channels (e.g., Email, SMS, PagerDuty, Slack via Pub/Sub).
- Name and Save: Give your alert policy a descriptive name.

Specific Error Types to Consider Alerting On:

Persistent "Failed to fetch RFD feed" errors.
"Discord webhook failed" errors (could indicate an invalid URL or Discord API issues).
"Firestore operation error" (read/write failures).
Any unhandled panics or critical errors logged by the application.

Cloud Logging automatically manages log retention according to configured policies.

Troubleshooting

Common Issues

Deals Not Appearing in Discord:
1. Check Cloud Scheduler: Go to Cloud Scheduler in the GCP Console. Verify the job (rfd-bot-trigger) status. Look at "Last run" and "Result". If it's failing, check its logs.
2. Check Cloud Run Logs: Go to Cloud Logging and filter for your rfd-discord-bot service. Look for any errors related to the RFD site, Firestore operations, or sending messages to Discord.
3. Discord Config: Ensure the DISCORD_APP_ID, DISCORD_PUBLIC_KEY, and DISCORD_BOT_TOKEN environment variables are correctly populated. If deals don't flow to a server, verify the server admin ran /rfd-bot-setup set.
4. Firestore Data: Check Firestore subscriptions and deals collections to verify logic flows.
Authentication/Permission Errors in Logs:
1. Cloud Run to Firestore: Double-check that the Cloud Run service's service account has the "Cloud Datastore User" (or "Firestore User") role.
2. Cloud Scheduler to Cloud Run (for private services): If your service is private, ensure the Scheduler's service account has "Cloud Run Invoker" permission. (Not applicable if --allow-unauthenticated was used).
Incorrect Environment Variables:
- Verify GOOGLE_CLOUD_PROJECT and DISCORD_WEBHOOK_URL are correctly set in the Cloud Run service's environment variables (check the "Revisions" tab of your service in Cloud Run, select the active revision, and look at "Variables").

How to Diagnose

This README provides a comprehensive guide for developers to understand, set up, deploy, and manage the Multi-Server RFD Hot Deals Discord Bot on Google Cloud.

Name		Name	Last commit message	Last commit date
Latest commit History 99 Commits
cmd		cmd
docs		docs
internal		internal
scripts		scripts
testdata		testdata
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
go.mod		go.mod
go.sum		go.sum
implementation_plan.md		implementation_plan.md
project.toml		project.toml
walkthrough.md		walkthrough.md

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Repository files navigation

RFD Hot Deals Discord Bot

🚀 How to Setup and Use

🌟 Features

🛠️ For Developers & Self-Hosting

Local Development

Clone Repository

Environment Setup (Go)

Environment Variables for Local Testing

Running Locally

Running Tests

Google Cloud Deployment (Detailed & Hand-holding)

Prerequisites

Enable APIs

Configure Cloud Build Service Account Permissions

Configure Google Cloud Build (project.toml)

Create Firestore Instance

Configure Environment Variables for Deployment

Deploy Cloud Run Service

Set up Cloud Scheduler

Configure IAM Permissions

Usage

Error Logging & Alerting

Cloud Logging

Critical Alerts (How to Set Up)

Troubleshooting

Common Issues

How to Diagnose

About

Uh oh!

Releases

Packages

Uh oh!

Uh oh!

Contributors

Uh oh!

Languages

Folders and files

Latest commit

History

Repository files navigation

RFD Hot Deals Discord Bot

🚀 How to Setup and Use

🌟 Features

🛠️ For Developers & Self-Hosting

Local Development

Clone Repository

Environment Setup (Go)

Environment Variables for Local Testing

Running Locally

Running Tests

Google Cloud Deployment (Detailed & Hand-holding)

Prerequisites

Enable APIs

Configure Cloud Build Service Account Permissions

Configure Google Cloud Build (project.toml)

Create Firestore Instance

Configure Environment Variables for Deployment

Deploy Cloud Run Service

Set up Cloud Scheduler

Configure IAM Permissions

Usage

Error Logging & Alerting

Cloud Logging

Critical Alerts (How to Set Up)

Troubleshooting

Common Issues

How to Diagnose

About

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Uh oh!

Contributors

Uh oh!

Languages

Packages