ChatGPT Plugin

miniDesk

Triage your support inbox

Brand: Unknown

Category: Pending
Primary Subcategory: Pending

Integration details

Description

miniDesk is a customer support inbox for live chat, email and web forms. This plugin lets the owner of a miniDesk workspace work that inbox from ChatGPT. What you can do: - See how the queue looks right now: open conversations by status, the oldest wait and the median time to a first reply. - Find the conversations that have waited longest for an answer. - Search conversations by the words customers used, order numbers, names, email addresses or dates, with counts by status and topic. - Read a conversation and look up a customer and their recent conversations. - Search your workspace's reference documents and read one in full. - Have miniDesk write a draft reply for your team to review, or send a reply after you have approved the exact text. - Leave an internal note for your team, or propose a new reference document for an owner to switch on. - See how many open conversations each teammate is assigned. Who it is for: support leads, founders and agents who use miniDesk and want answers about their inbox without opening the dashboard. Limits, on purpose: the plugin works in the one workspace you approve when you connect. It cannot close, assign, merge or delete conversations, and it cannot send anything in bulk: every reply goes to one conversation, after you agree to the text. Internal notes and your team's private notes about customers are never returned. Drafts and proposed documents reach nobody until a person approves them. You need a miniDesk account and must be an owner of the workspace you connect. You can disconnect at any time in Settings → MCP.

Integration type: Plugin
Verification status: Not applicable
Platform: ChatGPT
Category: Pending
Primary Subcategory: Pending
Secondary Subcategories: None listed
Brand: Unknown
Access: Account required
First tracked: 2026-10-22
Tool count: 13
Geography: US

Agent DiscoverabilityChatGPT Plugin Discovery Score

ChatGPT organic discovery is live

Your score is coming

ChatGPT now suggests Plugins on its own when they match a user's request.Your Plugin Discovery Score measures how often yours appears, and it will show here as soon as it’s ready.

ChatGPT recommends Airtable when a user asks it to update a CRM. — ChatGPT surfaces Airtable because it matches the user's request.

Primary Subcategory

Competitive lineup

No Primary Subcategory assigned yetOnce miniDesk has a Primary Subcategory, the Plugins and Connectors it competes with will appear here with their scores.

Tools

13 tools agents can invoke

Leave an internal note on one conversation for the team. The customer never sees it, and miniDesk's AI does not read it. Use it to record what you found or what should happen next.
add_internal_note
Write a draft reply for a person to review. It reaches nobody until a person sends it, and the result says so. It replaces any draft already waiting on that conversation. Use this rather than reply_ticket when you want miniDesk's own model to compose the answer from the workspace's documents.
create_draft_reply
One conversation and its public messages. Internal notes are never included.
get_ticket
One customer: who they are, how long they have been writing in, and their ten most recent conversations. Give their email, or the ticket_id of one of their conversations. The team's private notes about them are never included.
get_customer
How the support queue looks right now: open conversations by status, how long the oldest has been waiting, and the median time to a first reply over the last 30 days (null when too few conversations got one). Always includes as_of.
get_support_summary
How many open conversations each active member is assigned.
get_agent_workload
AI drafts waiting for a person to review, oldest first, on conversations that are still open. Each includes the draft text. A draft reaches nobody until a person sends it from the inbox.
list_pending_drafts
Conversations with no reply since the customer's last message, oldest first. Not simply status new: a ticket somebody opened and left is unanswered too.
list_oldest_unanswered
Propose a reference document for this workspace: a policy, a rule, a product note. It is saved switched off and an owner has to approve it before the AI can read it, so say that rather than telling your user it has been added.
write_knowledge
The full text of one reference document. Get its document_id by searching first.
read_document
Find conversations by what was written in them, by when they were opened, or both, with everything that matched counted beside them: how many, how they split by status, and what they were about. Use query to find a specific ticket (words the customer used, an order number, a name or an email address), and the dates for anything about a period, such as what customers have been reporting this month. Without dates it covers the last 30 days, or the last year when there is a query. Dates are UTC and both ends are included. The counts always describe everything that matched; the tickets are one page of it, most recently active first, so read the counts for the shape and the rows for the wording.
search_tickets
Find this workspace's own reference documents: policies, rules, product notes. Returns titles and ids, not the text. Search by topic, and call read_document with an id to read one. An empty query lists everything available.
search_knowledge
Send a reply to the customer, in your own words. This reaches a real person and cannot be unsent, so show the text to your user and get their agreement before calling it. One message to one conversation; there is no way to send in bulk. Use create_draft_reply instead when you want it reviewed first.
reply_ticket

FAQs

miniDesk ChatGPT Plugin FAQ

How the directory, categories and Discoverability Score work.

Read the methodology

How do I improve miniDesk's ChatGPT Plugin discoverability?

The levers are the listing surface agents actually read: names, descriptions, keywords, tool metadata, and registry health. Which lever matters depends on where discovery breaks, which is what continuous measurement shows.

Where is this profile measured?

This profile uses the geography attached to the latest public registry snapshot: US. Locale tags are intentionally omitted.