⚡ Usage

Example conversations showing how an AI assistant drives mcp-bugzilla against real Firefox/Mozilla bugs.

Once registered, you talk to your assistant in plain language — it picks the right tool and fills in the parameters. Here are representative exchanges.

🔍 Reading bugs

You: "Show me bug 1234567"
AI: Fetches and displays the bug summary, status, assignee, and metadata.

You: "Search for open Firefox crashes with priority P1"
AI: Queries product=Firefox, status=NEW|ASSIGNED, priority=P1, severity=critical.

You: "Show me the change history for bug 500000"
AI: Displays the full audit trail of field changes with timestamps.

💬 Discussion

You: "What comments are on bug 1800000?"
AI: Lists all comments with author names and timestamps.

You: "Mark bug 9999 as RESOLVED FIXED with a comment explaining the fix"
AI: Updates status/resolution and posts the comment in one operation.

📝 Filing and editing

You: "File a bug for a Firefox crash on macOS when opening PDFs"
AI: Creates the bug in the DOM: PDF Viewer component with the right fields.

You: "Who is assigned to the most recent WebGL regression?"
AI: Searches for WebGL regressions and shows assignee details.

📎 Attachments

You: "Attach this patch to bug 1234567 and request review from alice@mozilla.com"
AI: Uploads the base64-encoded patch and sets a review? flag for the requestee.

You: "List the attachments on bug 1800000"
AI: Returns each attachment with filename, type, and patch/obsolete status.

📦 Products, users, and fields

You: "What components does the Firefox product have?"
AI: Fetches product details and lists all components.

You: "Who is lsalzman@mozilla.com on Bugzilla?"
AI: Returns user profile including display name and real name.

You: "What are the valid values for bug severity?"
AI: Queries get_bug_fields for severity and lists all legal values.

See the Tools reference for every parameter each of these calls accepts.