Niches · Customer support & helpdesk
Summarise tickets and report on the queue
Four or more companies' docs show a full fit.
What counts as done
A summary that names the open questions, or a category table with counts, produced from paged calls without exceeding the rate limit, and written back as a note or left outside the helpdesk.
Measured, for our own tests: the summary is under 200 words. No docs we read state these figures, so they don't count against a listing here.
How agents ask for it: summarise this thread
, what are the top issues this week
, categorise our ticket history
, are we falling behind on tickets
Does the job 7
The docs show it meets the test above, and nothing we read says otherwise.
| Listing | Why |
|---|---|
| Gorgias API + MCP C Gorgias Inc. | get_reporting_stats and ticket_pattern_breakdown group tickets by contact reason, messages carry stripped_text, and add_internal_note writes back. source |
| HelpDesk D Text, Inc. | About thirty report endpoints cover new tickets, status, response and resolution time, tickets page by cursor, and a private message writes the summary back. Calls are metered against the Text plan. source |
| Intercom API + MCP BB Intercom, Inc. (with Intercom R&D Unlimited Company in Ireland) | Conversations page by cursor, reporting data exports run as jobs, and add_internal_note writes the summary back without reaching the customer. source |
| Kustomer D Kustomer, LLC | Search with field selection pages conversations, AI summaries can be fetched in batches, and notes write a summary back. source |
| Plain API + MCP B Not Just Tickets Ltd | Threads page by cursor with chosen fields, and notes write the summary back; the MCP server suggests precisely this summary prompt. source |
| Pylon API + MCP C Pylon Labs, Inc. | Issues and their messages page by cursor, search narrows by filter, and /issues/{id}/note writes the summary back as an internal note. source |
| Zendesk Support API B Zendesk, Inc. | Incremental ticket export and search export page to completion, search/count gives counts, and an internal note carries the summary back; acc-12 runs such a monitoring agent. source |
Does the job, with a limit 14
It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.
| Listing | Why |
|---|---|
| Chatway MCP C Poptin Ltd | The insights skill pages list-conversations at up to 100 and reads messages by cursor, with add-note for write-back, but it warns of sampling and pagination truncation, and Free keeps 30 days. source |
| Chatwoot API D Chatwoot Inc. | Paged conversation lists, messages and report endpoints (counts by inbox, label matrix) exist, and a private note can carry the summary, but cloud rate limits are unpublished and records have no field selection. source |
| Crisp API + MCP D Crisp IM SAS | Conversations and messages page through REST and a note carries the summary back, but pages stop at 50 and the daily token quotas limit how many conversations one run can read. source |
| Dixa D Dixa ApS | Messages, notes and activity logs are readable and a note writes back, but paging uses opaque pageKey at 10 requests a second per token, slow for thousands of conversations. source |
| Drag E DragApp.com Ltd | get_thread reads a thread, analytics tools give response times and daily activity, and add_comment writes back; how far list_threads pages was not documented. source |
| FreeScout F FreeScout | Conversations page up to 100 at a time with threads embedded, Get Report gives queue figures, and a note thread writes back; no rate limits are documented, so pacing depends on the operator's server. source |
| Freshdesk API + MCP D Freshworks Inc. | Conversations return body_text and a private note carries the summary, but ticket lists stop at 300 pages and closed tickets archive after 120 days of inactivity. source |
| Front API + MCP BB FrontApp, Inc. | read_conversation returns the whole timeline and read_message strips quoted replies; add_comment writes the summary back. Bulk counts mean paging at 50 requests a minute on Starter. source |
| Gladly C Gladly Software, Inc. | The Export API sends conversations as daily files and a note can carry a summary back, but files are removed after 14 days and reports allow 10 calls a minute. source |
| Help Scout API + MCP B Help Scout PBC | Reports endpoints (volumes, productivity, drilldowns) and paged conversation lists cover the queue, and Create Note writes back, but reports need Plus or Pro and each v3 search counts double. source |
| LiveAgent D Quality Unit, s. r. o. | get_ticket_summary returns an AI summary on personal connections and add_note writes back, but v3 GET /tickets returns at most 10,000 tickets per filter and rate-limit responses are undocumented. source |
| Missive API + MCP C Heliom Inc. | Analytics reports, message and comment lists, and posts for writing a summary back exist, but message bodies come only from Get a Message, one at a time. source |
| Zammad E Zammad GmbH | POST /tickets/{id}/summarize returns a summary, but acc-29 hit JSON parse errors on German quotation marks and acc-30 truncation from a hardcoded 1,024 max_tokens, both open. Paging has hard per-page limits. source |
| Zoho Desk B Zoho Corporation Private Limited | Threads, metrics and counts are readable and a private comment carries the summary back, but lists page 50 at a time and deep offsets cost up to 100 credits a call against a daily allowance. source |
Doesn't do this job 1
Listings in the category that do something else, with the reason.
| Listing | Why |
|---|---|
| Landbot MCP server E Landbot | No conversation, transcript or reporting tool is in the tool list. source |
What stops agents finishing
- model output rejected by the helpdesk's parser or truncated by a fixed token limit
- pagination caps below the number of tickets wanted
- each v3 search counts double against the limit
Around it
Before: support.research-context. After: knowledge.write-article, messaging.slack-notify.
Where this comes from
Weight 12 of 100 in Customer support & helpdesk: 6 of 42 accounts (acc-11, 12, 13, 29, 30, 34); read-only MCP vendors name this as their first use case
The same as data: /api/v1/use-cases/support.summarise-and-report.json, or the MCP tool get_use_case with id support.summarise-and-report. Other jobs in this category: /api/v1/niches?category=support.
For companies
Do agents find, use and choose your tools?
An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.