monday-code-ops

monday-code-ops

Debug, monitor, and operate monday code apps after deployment. This skill covers everything needed to work with a live monday code app: viewing logs, getting deployment URLs, managing environment variables and secrets, retrieving database connection strings, monitoring alerts, and troubleshooting production issues.

When to Use

• User wants to view or stream production logs
• User wants to get the deployment URLs (CDN or serverless hosting URL)
• User wants to check deployment status
• User wants to manage environment variables or secrets in production
• User wants to get the MongoDB connection string for direct database access
• User wants to check or configure alerts on the alerts board
• User wants to debug a production issue
• User wants to check cron job status or run a job on demand
• User wants to get a security scan report

Usage Examples

/monday-code-ops logs
/monday-code-ops status
/monday-code-ops env
/monday-code-ops secrets
/monday-code-ops db
/monday-code-ops alerts
/monday-code-ops urls

Or conversationally: "Show me the production logs", "What's the deployment URL?", "Get my MongoDB connection string", "Check the alerts board", "Why is my app failing in production?".

Prerequisites

• mapps CLI installed (npm -g i @mondaycom/apps-cli) and authenticated (mapps init)
• MONDAY_APP_ID environment variable set (find it in Developer Center > General Settings)
• App must be deployed at least once to monday-code

Verify prerequisites:

mapps --version
echo "MONDAY_APP_ID=${MONDAY_APP_ID}"

If MONDAY_APP_ID is not set, ask the user. They can find it in the Developer Center URL or General Settings page.

Instructions

1. Get Deployment Status

Check the current deployment status and version info. See references/deployment-status.md for full CLI flags and MCP alternatives.

mapps app-version:list -i ${MONDAY_APP_ID}
mapps code:status -i <version_id>

2. Get Deployment URLs

Backend (Serverless) URL: Retrieve via mapps code:status -i <version_id> or the monday SDK context (monday.get("context") returns appVersion.mondayCodeHostingUrl).

Frontend (CDN) URL: Available in the Developer Center > App > Host on monday > Client-side code section.

See references/deployment-status.md for MCP alternatives.

3. View Logs

Stream or search production logs. See references/logs.md for full CLI flags, date filtering, and regex search options.

mapps code:logs -i <version_id>                          # Stream live logs
mapps code:logs -i <version_id> -t http                  # HTTP request logs only
mapps code:logs -i <version_id> -t console               # Console/stdout only
mapps code:logs -i <version_id> -s live -r "error|timeout"  # Search with regex

4. Environment Variables

Manage environment variables. See references/env-and-secrets.md for full CLI flags and MCP alternatives.

mapps code:env -i ${MONDAY_APP_ID} -m list-keys           # List keys
mapps code:env -i ${MONDAY_APP_ID} -m set -k KEY -v "val" # Set a variable

Environment variables require a re-deploy to take effect.

5. Secrets Management

Manage secret variables. See references/env-and-secrets.md for full CLI flags, common secrets, and runtime access patterns.

mapps code:secret -i ${MONDAY_APP_ID} -m list-keys              # List keys
mapps code:secret -i ${MONDAY_APP_ID} -m set -k KEY -v "secret" # Set a secret

6. Database Connection String

Get the MongoDB connection string. See references/database.md for limits and important notes.

mapps database:connection-string -a ${MONDAY_APP_ID}

MNDY_MONGODB_CONNECTION_STRING is auto-injected after first deploy — never set it manually.

7. Alerts and Monitoring

monday code alerts auto-create a monday.com board for notifications. See references/alerts-and-monitoring.md for detailed setup, alert types, and best practices.

3 alert types: HTTP error rate, HTTP latency response, and runtime limit quota. Set up via Developer Center > Host on monday > Server-side code > Alert policies tab.

Query the alerts board via MCP:

mcp__monday__get_board_items_page({ boardId: ALERT_BOARD_ID })

8. Cron Jobs Management

View and manage scheduled background jobs. See references/cron-jobs.md for full CLI flags and constraints.

mapps scheduler:list -a ${MONDAY_APP_ID}                     # List jobs
mapps scheduler:create -a ${MONDAY_APP_ID} -n "name" -s "0 * * * *" -e "endpoint"  # Create
mapps scheduler:run -a ${MONDAY_APP_ID} -n "name"            # Run on demand

Cron routes MUST use /mndy-cronjob/ prefix. Max 5 jobs per region. IL region does not support cron.

9. Security Reports

Get security scan report. See references/security-and-storage.md for full CLI flags.

mapps code:report -i <version_id>

10. Storage Data Export

Export stored data for a customer account. See references/security-and-storage.md for full CLI flags and MCP alternatives.

mapps storage:export -a ${MONDAY_APP_ID} -c <ACCOUNT_ID>

11. Version Management

Manage app versions, promotions, and manifests. See references/version-management.md for full CLI flags and MCP alternatives.

mapps app-version:list -i ${MONDAY_APP_ID}               # List versions
mapps app:promote -a ${MONDAY_APP_ID} -i <version_id>    # Promote to live
mapps manifest:export -a ${MONDAY_APP_ID} -p ./exports   # Export manifest

Examples

Example 1: Debug a failing production app

User says: "My app is returning errors in production"

Actions:

1. Get the version ID: mapps app-version:list -i ${MONDAY_APP_ID}
2. Check deployment status: mapps code:status -i <version_id>
3. Stream console logs: mapps code:logs -i <version_id> -t console
4. Search for errors: mapps code:logs -i <version_id> -s live -r "error|exception|failed"
5. Check env vars are set: mapps code:env -i ${MONDAY_APP_ID} -m list-keys

Result: Found uncaught exception in webhook handler. Missing MONDAY_SIGNING_SECRET — set the secret and re-deployed to fix.

Example 2: Get database access for debugging

User says: "I need to look at the data in my production database"

Actions:

1. Get the connection string: mapps database:connection-string -a ${MONDAY_APP_ID}
2. Wait 1-2 minutes for the connection string to activate
3. Connect with MongoDB Compass or mongosh using the returned URI

Result: Provided connection string. User connected via MongoDB Compass and inspected the tasks collection.

Example 3: Set up monitoring for a new deployment

User says: "I just deployed, how do I monitor it?"

Actions:

1. Check deployment status: mapps code:status -i <version_id>
2. Guide user to set up alerts via Developer Center > Alert policies
3. Show how to query the alert board: mcp__monday__get_board_items_page({ boardId: ALERT_BOARD_ID })
4. Stream initial logs to verify healthy operation: mapps code:logs -i <version_id>

Result: Deployment confirmed healthy. Alert policies created for HTTP error rate (>5%) and latency (>2000ms at P95).

Troubleshooting

"My app is returning errors in production"

Cause: Application error in deployed code — could be missing env vars, unhandled exceptions, or dependency issues. Solution: Follow Example 1 above. Check logs first (-t console for app errors, -t http for request errors), then verify env vars and secrets are set.

"My app is slow / timing out"

Cause: Slow database queries, external API timeouts, or resource limits. Solution: Check HTTP latency logs (-t http), search for timeout patterns (-r "timeout|slow|ETIMEDOUT"), review cron jobs consuming resources (scheduler:list), and check the security report for dependency issues.

"Environment variable or secret not working"

Cause: Env vars require re-deploy, or wrong accessor used in code. Solution: Verify keys are set (code:env -m list-keys / code:secret -m list-keys). For multi-region apps, check the correct region with -z. Re-deploy after changes. In code, use EnvironmentVariablesManager for env vars and SecretsManager for secrets (both from @mondaycom/apps-sdk).

"MNDY_MONGODB_CONNECTION_STRING is undefined"

Cause: Auto-injected after first deploy — won't exist if the app has never been deployed. Solution: Deploy at least once. Verify with code:env -m list-keys. Get it directly with database:connection-string. For local dev: docker run -d -p 27017:27017 mongo:7 and set manually in .env.

"Deployment failed or stuck"

Cause: Build errors, auth issues, or incorrect app ID. Solution: Check status (code:status), verify auth (mapps app:list), verify app ID in Developer Center, build locally first (npm run build), ensure node_modules/ is excluded.

Notes

• Always use MCP tools (mcp__monday-apps__*) when available for programmatic access
• Multi-region apps need per-region secret/env var configuration
• Alerts are global (apply to all regions), but cron jobs are per-region
• Logs require a version ID — get it via mapps app-version:list first
• After changing env vars or secrets, a re-deploy is required
• Use --verbose on any command for advanced debug logging