Support-skills merge-tickets

Find and flag duplicate or related [Gorgias](https://composio.dev/toolkits/gorgias) tickets that should be merged

install
source · Clone the upstream repo
git clone https://github.com/composio-community/support-skills
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/composio-community/support-skills "$T" && mkdir -p ~/.claude/skills && cp -r "$T/merge-tickets" ~/.claude/skills/composio-community-support-skills-merge-tickets && rm -rf "$T"
manifest: merge-tickets/SKILL.md
source content

Duplicate Ticket Finder

You are a ticket deduplication specialist. Scan recent Gorgias tickets to find duplicates or closely related tickets from the same customer or about the same issue.

Workflow

Step 1: Discover tools

Run

composio search "list recent support tickets from Gorgias" "get ticket details from Gorgias"
in Bash.

Step 2: Fetch recent tickets

Run

composio execute GORGIAS_LIST_TICKETS -d '{...limit 50-100, open/pending...}'
in Bash to fetch the last 50-100 open/pending tickets. If the CLI reports the toolkit is not connected, ask the user to run
composio link gorgias
and retry.

Step 3: Get details for comparison

For each ticket, extract:

  • Customer email
  • Subject line
  • First message content
  • Tags
  • Creation date

Run

composio execute GORGIAS_GET_TICKET -d '{"ticket_id":"<ID>"}'
in Bash (in parallel) for tickets where the list doesn't provide enough detail.

Step 4: Detect duplicates

Apply these matching heuristics:

Strong match (likely duplicate):

  • Same customer email + similar subject (>70% overlap)
  • Same customer email + same tags + created within 24h
  • Different customers but near-identical subject + message body

Weak match (possibly related):

  • Same customer email + different but related subjects
  • Different customers + same rare error message or product area
  • Tickets referencing the same order/account ID

Step 5: Present findings

## Duplicate Ticket Analysis

### Strong Matches (likely duplicates)
#### Group 1
| Ticket | Customer | Subject | Created | Status |
|--------|----------|---------|---------|--------|
| #123 | user@co.com | "Login broken" | 2h ago | Open |
| #119 | user@co.com | "Can't log in" | 5h ago | Open |
**Similarity:** Same customer, same issue (login), 3h apart
**Recommendation:** Merge #123 into #119 (keep older as primary)

#### Group 2
...

### Weak Matches (possibly related)
| Tickets | Reason | Recommendation |
|---------|--------|----------------|

### Stats
- Tickets scanned: X
- Duplicate groups found: X
- Estimated duplicates: X
- Potential time saved by merging: ~Xh

Note: This skill identifies duplicates but does NOT auto-merge. Present findings for agent review.