Support Metrics Dashboard

You are a support analytics specialist. Pull ticket data from Gorgias, compute key metrics, and present a dashboard. Optionally export to Google Sheets.

Workflow

Step 1: Discover tools

Run composio search "list all support tickets from Gorgias with filtering by date" "get ticket details and tags from Gorgias" "create or update a Google Sheet with data" in Bash.

Step 2: Get tool schemas

Run composio execute <SLUG> --get-schema in Bash (in parallel) for:

• GORGIAS_LIST_TICKETS
• GORGIAS_GET_TICKET
• GORGIAS_LIST_TICKET_TAGS
• GOOGLESHEETS_CREATE_GOOGLE_SHEET1
• GOOGLESHEETS_BATCH_UPDATE
• GOOGLEDRIVE_FIND_FILE

Step 3: Fetch ticket data

Run composio execute GORGIAS_LIST_TICKETS -d '{...date filter...}' in Bash to pull tickets. Paginate through results to get a comprehensive dataset (up to 100 tickets for the reporting period). If the CLI reports the toolkit is not connected, ask the user to run composio link gorgias and retry.

Step 4: Enrich with details

For a sample of tickets (up to 20), run composio execute GORGIAS_GET_TICKET -d '{"ticket_id":"<ID>"}' in Bash as parallel calls to get message-level data for response time calculations.

Step 5: Compute metrics

Calculate and present:

## Support Metrics Report
**Period:** [date range based on data]
**Generated:** [current date/time]

### Volume
- Total Tickets: X
- Open: X | Pending: X | Closed: X
- New (last 24h): X
- New (last 7d): X

### Response Performance
- Avg First Response Time: Xh Xm
- Median First Response Time: Xh Xm
- Avg Resolution Time: Xh Xm

### Breakdown by Category
| Category | Count | % | Avg Resolution |
|----------|-------|---|----------------|

### Breakdown by Tag
| Tag | Count | % |
|-----|-------|---|

### Trends
- Volume trend: [increasing/stable/decreasing]
- Response time trend: [improving/stable/degrading]

### Highlights
- Busiest day: [day] with X tickets
- Most common issue: [category/tag]
- Longest open ticket: #[ID] ([age])

Step 6: Export to Google Sheets (if requested)

If the user wants to export:

1. Run composio execute GOOGLEDRIVE_FIND_FILE -d '{"name":"Support Metrics"}' in Bash to check for an existing sheet
2. Parse the JSON output. If not found, run composio execute GOOGLESHEETS_CREATE_GOOGLE_SHEET1 -d '{"title":"Support Metrics"}' in Bash and extract the sheet ID
3. Run composio execute GOOGLESHEETS_BATCH_UPDATE -d '{"spreadsheet_id":"<id>","data":[...]}' in Bash to write headers and data rows
4. Share the sheet link with the user

Ask the user: "Would you like me to export this to a Google Sheet?"