Open Source AI4 мин. чтения

SkyEagle888/coding-agent-monitor: Automated Tracking for AI Tool Updates

The coding-agent-monitor automatically tracks new versions of AI development tools and sends Discord alerts. Runs on GitHub Actions with no additional servers required.

SkyEagle888/coding-agent-monitor: Automated Tracking for AI Tool Updates
Illustration of a robot working with data and graphs.
AI tool update monitoring dashboard interface

Key takeaways

  • coding-agent-monitor is an open-source tool for tracking AI tool updates via GitHub Actions
  • Daily version checks at 01:00 UTC with Discord notifications
  • Monitors 4+ popular CLI tools including Gemini CLI and Qwen Code
  • Free service requiring no additional server deployment
  • Maintains history of last 5 versions per tool in RELEASES.md
  • Interactive monitoring dashboard with mobile-responsive design

What is coding-agent-monitor?

coding-agent-monitor is an automated solution for tracking updates to AI development tools. Running as a GitHub Actions workflow, it eliminates the need for separate server configuration. Its primary function is alerting developers about new CLI tool versions via Discord.

System architecture

The solution consists of several core components:

  • Python script monitor.py for repository checks
  • Configuration file watchlist.json listing monitored tools
  • Auto-updated versions.json file
  • Interactive dashboard/index.html for data visualization

Technical requirements

System requirements include:

  • GitHub repository with enabled Actions
  • Python 3.10+ for local development
  • Discord server for notifications
  • Approx. 5MB storage for version history

How the monitoring system works

The update process runs daily at 01:00 UTC via GitHub Actions. The system checks each repository from watchlist.json for new releases. When changes are detected:

  • A Discord message with update details is generated
  • versions.json file updates (stores last 5 versions)
  • Entries are added to RELEASES.md and CHANGELOG.md
  • Interactive dashboard rebuilds

Update check process

The monitoring algorithm includes:

  1. Fetching repositories from watchlist.json
  2. Querying GitHub API for each repository
  3. Comparing latest known release with current
  4. Processing changes (if any)
  5. Saving updated data
  6. Sending notifications

Scheduling adjustments

Users can modify check frequency by editing the cron expression in .github/workflows/monitor.yml. Default is '0 1 * * *' (daily at 01:00 UTC). For active projects, we recommend checking every 6 hours: '0 */6 * * *'.

Supported tools

By default, the system tracks 4 key AI tools:

  • Google's Gemini CLI (google-gemini/gemini-cli)
  • Qwen Code CLI by QwenLM (QwenLM/qwen-code)
  • OpenCode by anomalyco (anomalyco/opencode)
  • Oh-My-OpenCode by code-yeongyu (code-yeongyu/oh-my-opencode)

Tool selection criteria

Optimal repositories for monitoring:

  • Use standard GitHub releases with version tags
  • Maintain consistent update schedules
  • Are published as open-source projects
  • Follow clear versioning semantics

Adding custom tools

To monitor additional repositories, add entries to watchlist.json:

{
  "id": "your-tool",
  "owner": "owner-name",
  "repo": "repo-name",
  "emoji": "🚀"
}

Configuring Discord notifications

To enable alerts, create a Discord webhook and add its URL as a GitHub secret:

  1. Go to "Server Settings" → "Integrations" → "Webhooks"
  2. Create new webhook and select notification channel
  3. Copy webhook URL
  4. Add as DISCORD_WEBHOOK secret in GitHub repository settings

Notification format

Discord messages include:

  • Updated tool name
  • New version number
  • Publication date
  • Release link
  • Brief changelog
  • Emoji for visual distinction

Interactive monitoring dashboard

The project includes a web dashboard with:

  • Tool cards showing current version and release date
  • Timeline of last 5 updates
  • CHANGELOG.md table
  • Mobile-responsive design

Dashboard features

Users can:

  • Filter tools by update status
  • Compare versions across products
  • View complete change history
  • Export data to CSV

Local dashboard testing

To test the dashboard locally:

cd dashboard
python -m http.server 8080

Then open http://localhost:8080 in your browser.

Benefits

  • Free service — operates within GitHub Free tier
  • Automatic history — all changes logged in RELEASES.md
  • Easy setup — no DevOps knowledge required
  • Extensible — supports any GitHub repository
  • Open-source — MIT licensed for modifications

Comparison with alternatives

Feature coding-agent-monitor Alternatives
Cost Free Often paid
Setup 5 minutes Requires server
Flexibility Full Limited

Getting started

  1. Clone or fork SkyEagle888/coding-agent-monitor
  2. Configure Discord webhook and add to GitHub Secrets
  3. Enable workflow in repository's Actions section
  4. Add custom tools to watchlist.json if needed

Verification

After setup:

  1. Run workflow manually
  2. Check execution log
  3. Confirm test notification receipt
  4. Verify RELEASES.md updates

Questions & answers

How often does coding-agent-monitor check for updates?

By default, the system runs daily at 01:00 UTC. This schedule can be modified in .github/workflows/monitor.yml.

Can I add my own tool for monitoring?

Yes, simply add an entry to watchlist.json with the repository owner and name. The system supports monitoring any public GitHub project.

Where is version history stored?

History is maintained in two formats: structured (versions.json) and human-readable (RELEASES.md). versions.json stores the last 5 versions per tool.

Is there a cost to use coding-agent-monitor?

No, the project is completely free within GitHub Free tier limits (2000 workflow minutes/month).

How can I view update history conveniently?

Use the interactive dashboard in the dashboard folder, which displays current versions, change timeline, and detailed changelog.

Can I change the check frequency?

Yes, edit the cron expression in .github/workflows/monitor.yml. Use crontab.guru to generate proper schedule formats.

How are connection errors handled?

The system retries after 30 minutes for network/GitHub API errors. After 3 failures, Discord notifications are sent.

Are private repositories supported?

Yes, but requires setting up a personal access token with repo permissions in GitHub Secrets.