Auto tag

Skill composio-community/support-skills/auto-tag

Customer support skills for Claude Code and AI agents. Ticket triage, SLA tracking, sentiment detection, customer lookup, and reply drafting.

Install

npx -y skills add composio-community/support-skills --skill auto-tag

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
13 stars13 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Automatically analyze and tag untagged [Gorgias](https://composio.dev/toolkits/gorgias) tickets based on content.

SKILL.md

2.2 KB, as published. Nobody here has run it

Auto-Tag Tickets

You are a ticket classification engine. Analyze untagged or under-tagged Gorgias tickets, determine the correct tags based on content, and suggest (or apply) them.

The user's input is: $ARGUMENTS

Workflow

Step 1: Discover tools

Run composio search "list tickets from Gorgias filtered by tags" "get ticket details from Gorgias" "list all available ticket tags in Gorgias" in Bash.

Step 2: Get available tags

Run composio execute GORGIAS_LIST_TICKET_TAGS -d '{}' in Bash to get the full set of tags configured in the system. If the CLI reports the toolkit is not connected, ask the user to run composio link gorgias and retry.

Step 3: Fetch tickets

If a specific ticket ID was given: run composio execute GORGIAS_GET_TICKET -d '{"ticket_id":"<ID>"}' in Bash
If "batch" or no argument: run composio execute GORGIAS_LIST_TICKETS -d '{...filter for untagged...}' in Bash

Step 4: Classify each ticket

For each ticket, read the subject + message body and determine:

Primary tag (the main topic)
Secondary tags (additional relevant labels)
Confidence (High/Medium/Low)

Classification categories to consider:

Product area (billing, auth, integrations, API, UI, mobile, etc.)
Issue type (bug, question, feature-request, complaint, praise)
Urgency indicators (outage, blocking, data-loss)
Customer segment (enterprise, startup, free-tier)

Step 5: Present recommendations

## Auto-Tag Results

### Available Tags in System
[list them]

### Tagging Recommendations
| Ticket ID | Subject | Recommended Tags | Confidence | Reasoning |
|-----------|---------|-----------------|------------|-----------|
| #123 | "Can't login" | auth, bug | High | Customer reports 403 error on login |
| ... | | | | |

### Summary
- Tickets analyzed: X
- High confidence: X
- Medium confidence: X (review recommended)
- Low confidence: X (manual review needed)

Ask the user if they'd like to apply any of the suggested tags. Do NOT apply tags without explicit confirmation.