Menu
Mattering Automations connector

A Wistia MCP for the library and the numbers, not just the embed code

Connect one Wistia API token and 51 nodes land on the canvas: list and update media, move and tag it in bulk, import from a URL, manage folders and captions, read channels and webinars, change player customizations, and pull both the stats counters and the newer analytics surface. Video performance stops being a tab someone remembers to open and becomes a step in a workflow.

51 ready-made nodes One credential to connect Works with Claude, ChatGPT, and Matter

What the Wistia MCP connector does

The Wistia connector for Mattering Automations is 51 nodes on Wistia's modern API, pinned to the 2026-07 release: 34 read nodes and 17 action nodes. Media, folders and subfolders, the deleted media trash, captions, tags and bulk tagging, channels and channel episodes, webinars and their registrations, player customizations, account usage, and two separate reporting surfaces, the older stats counters and the newer analytics family with its timeseries, traffic breakdowns, embed locations and form conversions.

Every node speaks MCP, the open protocol for connecting AI assistants to tools, so Claude, ChatGPT, and any MCP-capable assistant can use them, alongside Matter, the AI assistant built into the platform. Reading is the large half here, 34 of the 51 nodes, which makes it a good assistant surface: an assistant can find a video with Wistia: List Media or Wistia: Search, read how it performed with Wistia: Get Media Analytics, see where it is embedded with Wistia: Get Media Embed Locations, and summarise it, without anyone opening the Wistia dashboard. Wistia also publishes an MCP server of its own, which is covered in the FAQ below rather than skipped over.

Two things about Wistia's own API shape are worth knowing before you design a workflow, because both cause silently wrong answers rather than errors. First, the two reporting surfaces do not agree on their date conventions: the analytics endpoints treat the end date as exclusive, so a range ending on the last day of a month omits that day, while the older stats by-date endpoints are inclusive. Second, the leaderboard style reads are ranked and truncated, not merely paged. Wistia: Get Account Top Content, Wistia: Get Media Embed Locations and Wistia: Get Media Traffic Breakdown sort rows and cut them off, and the embed locations read defaults to only ten rows, so a page you believed was not embedding a video may simply have fallen below the cut. That has a consequence worth stating plainly: those rows cannot be safely summed across chunks or stitched from several narrower requests, because an entity outside the cut in one sub period does not appear in it at all, so the total looks complete and under counts everything in the middle of the pack. Ask for the window you actually want, in one request.

Bulk writes here are asynchronous, and the response is an acknowledgement rather than a result: Wistia: Archive Media, Wistia: Restore Archived Media, Wistia: Move Media, Wistia: Bulk Tag Media, Wistia: Import Media From URL and Wistia: Restore Deleted Media all queue the work and hand back a job, which Wistia: Get Background Job Status then reports on, per item, with the error text for anything that failed. Nothing has happened when that first success arrives, so a workflow that writes and immediately reads gets stale data unless it polls. Four nodes are flagged destructive, and their blast radius is worth reading before you wire them: Wistia: Delete Media takes the video's captions, customizations and stats with it and every embed of it stops working immediately, Wistia: Delete Folder is documented as permanent without documenting what happens to the media inside, Wistia: Delete Tag removes that tag from every video it was applied to so any saved filter or rule matching on it stops matching, and Wistia: Delete Captions removes a language track that has to be re-ordered or re-uploaded.

One connection, three surfaces

  • Visual automations. Drag Wistia nodes onto the canvas and wire them to triggers, conditions, and other apps.
  • AI agents, yours or ours. These nodes speak MCP, the open protocol, so Claude, ChatGPT, and any MCP-capable assistant can use them, and Matter, the agent built into the platform, uses them natively.
  • The whole platform. Combine with native tasks, projects, and other connectors, no middleware in between.

Every Wistia node, ready to drop on the canvas

51 nodes, generated from the connector itself so this list always matches what ships. Expand a category to browse.

Read and query nodes 34

Wistia: Get Account

Get the account the API token belongs to: its name, its Wistia URL, and how many medias, folders and channels it holds. This is the cheapest call in the connector and the only one that works with any token permission level, which makes it the right way to confirm a credential works before trusting anything else.

Wistia: Get Account Analytics

Account-wide totals for one date range: unique loads, unique plays, plays, play rate, played time, unique visitors, engagement rate and form conversions. One object covering the whole window, not a series - Wistia: Get Account Analytics Timeseries is the series. The end date is EXCLUSIVE on this surface, so ask for the first of next month to include the whole of this one. The range may not exceed two years.

Wistia: Get Account Analytics Timeseries

Account-wide performance as a series: one row per day, week or month, each carrying unique loads, unique plays, plays, play rate, played time, unique visitors, engagement rate and form conversions. Any metric may come back null for a bucket with no activity. The end date is EXCLUSIVE and the range may not exceed two years. Bucket width is chosen per request, so the same node returns daily or monthly rows depending on what you ask for.

Wistia: Get Account Stats

Lifetime totals for the whole account: loads, plays and hours watched. Three numbers, no window, no parameters - the cheapest possible health check on a video library. For a series over time use Wistia: Get Account Stats By Date, and for rate metrics use the Analytics nodes. Needs a token with Read detailed stats, which is a separate permission from ordinary read access.

Wistia: Get Account Stats By Date

Account-wide loads, plays and hours watched broken out by day, between two dates inclusive. Leaving both dates blank gives yesterday and today, which makes this a cheap daily poll. Each row carries its own date, so a long backfill is ONE request rather than one per day.

Wistia: Get Account Top Content

The account's best-performing content over a date range, ranked: plays, play rate, engagement rate, played time and unique visitors per media, channel or folder, with each media's name, type, duration and thumbnail alongside. This is the leaderboard a weekly digest is built from. Because the rows are RANKED AND TRUNCATED, results from two different date ranges cannot be added together - re-request the range you want. Note the group_by value for folders is still spelled project, which the Stats API kept.

Wistia: Get Account Usage

Get the account's plan tier, whether uploads are currently allowed, and its media and storage usage against its limits. Use this as a guard in front of an upload workflow: data.can_upload false with data.upload_blocked_reason explains why before a bulk import fails halfway. Usage and limits are visible only to account owners and managers; every other token gets null for data.limits, which is a permission result rather than an error.

Wistia: Get Background Job Status

Check how a queued bulk operation is going. Every bulk write in this connector - archive, restore, move, bulk tag, import from URL, restore from trash - returns immediately with a job id and does the work afterwards, so this node is what turns an acknowledgement into an answer. Poll until data.background_job_status.status is finished, then read the per-item results array to see which media succeeded and which failed, and why.

Wistia: Get Folder

Get one folder's record: name, description, media count, public-access flags, and whether anonymous visitors may upload or download. Despite the path parameter being named id by the vendor, it wants the HASHED id, not the numeric one that appears in the response body.

Wistia: Get Folder Stats

Lifetime loads, plays and hours watched for every media in one folder. Note the path: when projects were renamed to folders, the Stats API path /stats/projects/{id} was deliberately NOT renamed, so this node sends a folder's hashed id to a projects path. That is the vendor's decision, not a mistake here. The number_of_videos field the endpoint returns is marked deprecated; read data.media_count from Wistia: Get Folder instead.

Wistia: Get Media

Get one media's full record: name, description, duration, processing status and progress, thumbnail, tags, the folder and subfolder it sits in, and the assets array with each rendered file's URL, dimensions, size and content type. Poll data.status until it reads ready after an import; anything else means the file is still being processed.

Wistia: Get Media Analytics

One video's totals for a date range: unique loads, unique plays, plays, play rate, played time, unique visitors, engagement rate, call-to-action impressions and conversions, and form conversions. This is the windowed counterpart to Wistia: Get Media Stats, which reports lifetime totals with no range. The end date is EXCLUSIVE and the range may not exceed two years.

Wistia: Get Media Analytics Timeseries

One video's performance as a series: a row per day, week or month carrying unique loads, unique plays, plays, play rate, played time, unique visitors, engagement rate, call-to-action impressions, conversions and conversion rate, and form conversions. This is the node a video dashboard is built on. Remember which columns are RATES: play rate, engagement rate and conversion rate are ratios and must never be added across rows, and the three unique counts cannot be added either. The end date is EXCLUSIVE; the range may not exceed two years.

Wistia: Get Media Customizations

Read every player setting saved against a media: colour, autoplay, which controls are shown, end-of-video behaviour, and the plugin block that holds captions, chapters, call-to-action and password settings. Note this payload is camelCase (playerColor, autoPlay) rather than snake_case like the rest of the modern API, because these keys are the player's own embed-option names. Values come back as strings even when they read like booleans.

Wistia: Get Media Embed Locations

Which pages a video is embedded on and how it performs on each: the domain, path, page title, loads, plays, play rate, played time, engagement rate and conversion rate per URL. This answers where a video is actually being watched, which is usually not where anyone remembers putting it. Ranked and truncated, defaulting to only 10 rows - raise per_page before concluding a page is not listed.

Wistia: Get Media Engagement

The engagement graph behind a video: the average share of the video watched, plus two arrays describing viewing across the video's length - engagement_data for how many viewers were still watching at each point, and rewatch_data for the parts people went back to. The arrays are positional buckets across the video, not timestamps, so a spike at index 4 means roughly four fifths of the way in for a five-bucket array. This is how a workflow finds the moment viewers drop off.

Wistia: Get Media Form Conversions

The actual people who filled in a video's Turnstile or email-gate form during a date range: first name, last name, email and when they converted. This is the lead list a video-to-CRM automation writes into a pipeline. Rows are at data.data[] inside an envelope and paging is cursor-only, from data.page_info.end_cursor. The end date is EXCLUSIVE; the range may not exceed two years.

Wistia: Get Media Stats

Lifetime performance for one video: loads, plays, play rate, hours watched, average engagement, unique visitors, and a breakdown of every call to action and annotation with its impression and click counts. Note play_rate and engagement here are RATES over the whole lifetime of the video, not sums, so they cannot be added to anything.

Wistia: Get Media Stats By Date

One video's loads, plays and hours watched broken out by day, between two dates inclusive. Leaving both blank gives yesterday and today. Each row carries its own date, so a ninety-day backfill is one request rather than ninety.

Wistia: Get Media Traffic Breakdown

Where a video's viewers came from, broken down by whichever dimension you pick: UTM campaign, source or medium, referring domain, or viewer screen size. Each row carries loads, plays and engagement rate. The DIMENSION COLUMN IS NAMED AFTER group_by, so a workflow reading these rows has to know which breakdown it asked for: group_by utm_source produces a utm_source column, not a generic one. Ranked and truncated.

Wistia: Get Webinar Analytics

How a webinar performed: registrations (including imported and on-demand ones), attendance, total and average watch time, engagement rate, chat and Q and A volume, and every poll question with its per-option response counts. Turn on include_post_event to add on-demand viewing after the live event ended, which is when the two post-event date fields start mattering; without that flag they are ignored.

Wistia: List Channel Episodes

List channel episodes across the account or within one channel, with each episode's title, summary, the media it wraps, whether it is published, and its scheduled publish time. Filter on published false to find episodes waiting to go live, which is the usual trigger for a publishing workflow.

Wistia: List Channels

List the account's channels: their names, how many media each holds, whether new episodes publish automatically, and whether the channel is set up as a podcast. A channel is Wistia's published video series; its contents are episodes, which Wistia: List Channel Episodes returns.

Wistia: List Deleted Media

List media that has been deleted but is still inside the account's restore window, with the folder each came from, when it was deleted, and - the useful one - when it stops being restorable. Wistia keeps deleted media for 30 days on most plans and 14 on free plans. A media missing from this list is gone for good.

Wistia: List Folders

List the account's folders with names, hashed ids, media counts and whether each is a shared folder or someone's personal library. Folders were called projects on the old v1 API and are what the Wistia UI calls folders today. Every folder picker in this connector reads its values from here, and note each folder carries BOTH a numeric id and a hashed id - almost everything wants the hashed one.

Wistia: List Media

List the account's media with names, hashed ids, durations, statuses, thumbnails, tags and the folder each one lives in. This is the node every other Wistia media node leans on: the hashed id it returns is what media pickers store. Filter by folder, media type, tags or archived state. The response is a bare array with no pagination wrapper, and each row carries its own cursor for cursor paging.

Wistia: List Media Captions

List every caption track on a media, with each track's language code, its English and native language names, whether it is still a draft, and the full SRT text. The text field is the whole transcript, so this doubles as the way to pull a video's transcript into a workflow that summarises or translates it.

Wistia: List Subfolders

List the subfolders inside one folder, in their display order. Subfolders are one level deep and belong to exactly one folder, which is why this node takes a folder and why the subfolder picker on Wistia: Move Media asks for the destination folder first.

Wistia: List Tags

List the account's tags with how many media each one is applied to. Tag names are the identifier here - there is no separate tag id - so the name from this node is what every other tag field wants.

Wistia: List Viewing Events

Individual viewing sessions: who watched what, when, from where, how much of it they saw, and any form conversion the session produced. This is the rawest data Wistia exposes and the input to lead-scoring workflows. Only the last two years of events are retained. Rows are one per session at irregular times, not per day, so this is a stream to filter rather than a series to chart.

Wistia: List Visitors

List the people who have watched something, with their visitor key, when they were last active, how many loads and plays they account for, whatever identity Wistia has captured (name, email, organisation) and their browser and platform. Most visitors are anonymous: filter on has_email to get only the ones a CRM workflow can act on.

Wistia: List Webinar Registrations

List who registered for a webinar and what they did: name, email, company, the UTM parameters they arrived with, whether and when they attended, how engaged they were, and their chat, poll and Q and A activity. This is the node a webinar-to-CRM workflow is built on. Filter on attendance to separate attendees from no-shows. Unlike most list endpoints here, the response is an envelope: rows are at data.registrations[] and paging is cursor-only, from data.page_info.end_cursor.

Wistia: List Webinars

List the account's webinars with titles, scheduled times and time zones, lifecycle and registration status, and the audience, host and panelist links. Webinars were called live stream events on the old v1 API; this is the same thing under the name the product uses. Filter on started to separate upcoming events from ones that have already run.

Wistia: Search

Search across the whole account in one call: media, folders, subfolders, channels, channel episodes and webinars. It searches TRANSCRIPTS as well as titles, so a query matching spoken words finds the video that said them. Results come back grouped by resource type under data.data, not as one flat list, so read data.data.medias[] rather than data[]. Needs a token with Read all data.

Action nodes 17

Wistia: Archive Media

Archive up to 100 media at once. Archiving takes a video out of the active library without deleting it, and Wistia: Restore Media puts it back, which makes this the reversible alternative to Wistia: Delete Media. Two kinds of media cannot be archived: webinar media, and Soapbox videos imported before 1 September 2023. Needs the Archiving feature on the account. Asynchronous: poll the returned job id.

Wistia: Bulk Tag Media

Apply a set of tags to a set of media in one call. Tags are APPENDED to whatever each media already carries, unlike Wistia: Update Media whose tags field replaces the lot, so this is the safe way to add a tag across a library. Tag names that do not exist yet are created. Asynchronous: poll the returned job id for the per-media outcome.

Wistia: Copy Media

Duplicate a media into a folder, optionally handing the copy to a different owner. The copy gets its own hashed id and its own stats starting from zero; the original is untouched. Note this is the ONE endpoint in the connector whose folder_id is the NUMERIC folder id rather than the hashed one, which the picker on that field handles for you.

Wistia: Create Captions

Add a caption track to a media from SRT text. Pass the SRT contents as a string; Wistia detects the language if you do not name one. This is the write half of a translation workflow: pull the existing track with Wistia: List Media Captions, translate the SRT, post it back under a new language code. Needs a token with Read, update and delete anything.

Wistia: Create Folder

Create a folder. Every field is optional, so an empty call makes an unnamed shared folder owned by the account owner. The response carries the new folder's hashed id at data.hashed_id, which is what an import or move node needs next. Note this endpoint's body keys are camelCase on the wire while the response is snake_case; the node handles that translation.

Wistia: Create Tag

Create a tag on the account without applying it to anything. Wistia: Bulk Tag Media creates tags implicitly as it applies them, so reach for this only when a tag needs to exist before any media carries it. A name already in use returns 422.

Wistia: Delete Captions

DESTRUCTIVE. Remove one language's caption track from a media. The track's timecoded text is gone; a purchased or hand-corrected transcript has to be re-ordered or re-uploaded to get it back, and any translation derived from it goes with it. The media itself and its other caption languages are untouched. Returns 200 with no body.

Wistia: Delete Folder

DESTRUCTIVE. Delete a folder. Wistia states the folder is permanently deleted and cannot be undone, and does NOT document what becomes of the media and subfolders inside it, so assume they go with it: treat this as deleting the folder's entire contents, along with every embed of every video in it. Move the media out first with Wistia: Move Media if you only want the folder gone. Check data.media_count on Wistia: Get Folder before running this.

Wistia: Delete Media

DESTRUCTIVE. Delete a media and everything attached to it: its captions, its customizations, its stats history and its channel episode if it has one. Every embed of this video on every page stops working immediately. Wistia's own reference for this endpoint states the deletion is permanent and cannot be undone, while the trash surface says deleted media stays restorable for 30 days on most plans and 14 on free plans - check Wistia: List Deleted Media before assuming either. To take a video out of circulation reversibly, use Wistia: Archive Media instead. Needs a token with Read, update and delete anything.

Wistia: Delete Tag

DESTRUCTIVE. Delete a tag from the account. The tag is removed from EVERY media it was applied to, so any saved filter, automation condition or channel rule that matched on it stops matching. The media themselves are not touched. Wistia states this cannot be undone; re-creating a tag of the same name does not re-apply it to anything.

Wistia: Import Media From URL

Pull a publicly reachable media file into Wistia by URL. This is how an automation gets video INTO Wistia: the connector does not use Wistia's separate multipart upload host, so the file must already be somewhere Wistia can fetch it. Imports from vimeo.com and wistia.com are refused by the vendor. The call returns a queued background job rather than the finished media, so follow it with Wistia: Get Background Job Status and then Wistia: Get Media until status reads ready. Leaving the folder blank makes Wistia create a NEW folder rather than using a default one.

Wistia: Move Media

Move up to 100 media into a folder, and optionally into a subfolder inside it. THIS ENDPOINT HAS ITS OWN RATE LIMIT of 10 requests per 5 minutes, separate from and far tighter than the account's 600 per minute, so move media in batches rather than one node run per video. Asynchronous: poll the returned job id. A partial failure comes back as HTTP 207 with a partial_errors array naming which media failed.

Wistia: Restore Archived Media

Bring archived media back into the active library, up to 100 at a time, placing them in a folder you name. This undoes Wistia: Archive Media; it does NOT recover deleted media - that is Wistia: Restore Deleted Media, a different surface with its own time limit. Asynchronous: poll the returned job id. Needs the Archiving feature on the account.

Wistia: Restore Deleted Media

Recover deleted media from the trash, up to 1000 at a time, either back where each one was or into a folder you name. Only works while the media is still inside its restore window - check Wistia: List Deleted Media and its restore_expires_at first. Asynchronous: poll the returned job id. This is a different surface from Wistia: Restore Archived Media, which un-archives rather than un-deletes.

Wistia: Update Folder

Rename a folder, rewrite its description, or change its sharing posture: whether it is publicly accessible and whether anonymous visitors may upload into it or download from it. Turning public access on exposes every media in the folder to anyone with the link, so treat it as a disclosure change rather than a cosmetic one. Fields left blank are not touched.

Wistia: Update Media

Rename a media, rewrite its description, replace its tag set or swap its thumbnail. Note that tags REPLACE the media's existing tags wholesale rather than adding to them - to add tags without losing the current ones, use Wistia: Bulk Tag Media, which appends. Fields left blank are not touched. Needs a token with Read, update and delete anything.

Wistia: Update Media Customizations

Change a media's player settings. The body is an open key-value document rather than a fixed field list: a non-null value creates or replaces that setting, and an explicit null DELETES it, falling the player back to the account default. Keys you do not send are left alone. Keys are camelCase (playerColor, autoPlay, videoFoam) - this surface did not move to snake_case with the rest of the API. Read the current settings first with Wistia: Get Media Customizations if you intend to preserve anything.

Connecting Wistia takes one credential

Wistia authenticates with a bearer token created in Wistia account settings under API access. The token belongs to the account rather than to a folder or a project, and it is shown once at creation, so copy it then. What decides whether a given node works is not a scope list but a permission level chosen when the token is created, and the levels are not nested the way people assume. Basic account reads such as Wistia: Get Account work with any valid token. The library reads, media, folders, captions, channels and episodes, need read access to folder and media data. Wistia: Search, Wistia: List Tags and the webinar reads need read access to all data. Every stats and analytics node needs Read detailed stats, which is a separate permission that ordinary read access does not imply, and which is the single most common surprise here: a token that reads your media list perfectly can return a permission error on every performance node. Write nodes need read, update and delete access. Mattering also pins Wistia's API version header on every request, because omitting it serves whatever release is current and a running automation would change shape as Wistia ships new ones. Feature gating exists alongside the permissions: Wistia documents archiving as needing access to the Archiving feature, and webinars, channels and detailed stats are features to confirm on your plan rather than assume. Wistia's own pricing and support pages are the place to check.

1

Sign in to Wistia, open your account settings, and go to the API access section. Create an API token and copy the value while it is shown.

2

Set the permission level to cover what you intend to automate. If any part of the workflow reads performance data, the token needs Read detailed stats specifically, not just read access to media.

3

In Mattering Automations, open Connectors, click Add Connector, choose Wistia, and create a credential with that token.

4

Click Test Connection. Mattering calls Wistia's account endpoint, which needs no parameters and no ids and, uniquely here, requires no permission level beyond a valid token, so a failure there is unambiguously an auth problem rather than a permission one.

5

All 51 Wistia nodes appear in the canvas palette. Run Wistia: List Media first, then a stats node, because those two together tell you whether the token's permission level covers the half of the connector you actually came for.

Common uses for the Wistia MCP

Weekly video performance without anyone opening the dashboard

Wistia: Get Account Analytics and Wistia: Get Account Analytics Timeseries give plays, watch time and engagement over a window, and Wistia: Get Media Analytics does the same per video. A scheduled workflow can write the series into a Mattering database, compare it against the previous period, and open a task only when something moved further than it should have. Two details make the report correct rather than plausible: the analytics end date is exclusive, so ask for one day past the last day you want, and engagement and play rate are rates rather than counts, so combining periods means weighting them by the loads and plays underneath rather than averaging the percentages.

Keep a growing library tidy on a schedule

Wistia: List Media, Wistia: List Folders and Wistia: Search find what has drifted, and Wistia: Move Media, Wistia: Update Media and Wistia: Bulk Tag Media put it back where it belongs. A sweep can find media untouched for a year and archive it with Wistia: Archive Media, or find videos missing a required tag and add it. Two behaviours shape the design: bulk writes are queued rather than immediate, so poll Wistia: Get Background Job Status before the next step assumes the work is done, and the move endpoint carries its own tight rate limit that is far lower than the account limit, so a loop moving one video per run will hit it quickly. Move in batches instead.

Publish a video and finish the checklist automatically

Wistia: Import Media From URL brings a rendered video in from wherever it was produced, and the rest of the launch checklist becomes steps rather than reminders: Wistia: Update Media sets the title and description, Wistia: Bulk Tag Media applies the campaign tags, Wistia: Create Captions attaches the transcript, and Wistia: Update Media Customizations applies the player settings the brand expects. One argument is worth filling in every time: an import with no folder named creates a new folder rather than using a default, so a scheduled import loop left unconfigured will litter the account with folders.

Find out where your videos actually live

Wistia: Get Media Embed Locations lists the pages a video is embedded on with the plays each one produced, and Wistia: Get Media Traffic Breakdown groups viewing by source, so a workflow can answer which page or which campaign is doing the work. Read the numbers with the ranking in mind: this response defaults to a small number of rows and is sorted, so raise the page size before concluding a page is not embedding the video, and do not assemble a picture of a quarter by stitching together twelve weekly requests, because each request ranks and cuts independently.

Captions and accessibility as a workflow step

Wistia: List Media Captions shows which languages a video has, Wistia: Create Captions adds a track, and Wistia: Delete Captions removes one. A scheduled audit can list the media in a folder, check each one for the languages your policy requires, and open a task for anything missing rather than relying on someone to notice. The delete node is flagged for a reason worth stating: a purchased or hand corrected transcript has to be re-ordered or re-uploaded, and translations derived from it go too.

Webinars and channels, read where the rest of the work happens

Wistia: List Webinars and Wistia: List Webinar Registrations bring the live event side into a workflow, so a registration can create a CRM record, a task or a follow up sequence in the same run, and Wistia: Get Webinar Analytics reports on the event afterwards. Wistia: List Channels and Wistia: List Channel Episodes cover the published collections. These are reads only in this connector: channel, episode and webinar authoring stays in Wistia, and what comes out here is what a workflow reacting to a webinar or a new episode actually needs.

Example: A Monday video report that a person only reads when it matters

Built on the Automations canvas in a few minutes, or described in plain language to Matter and built for you.

1

Trigger: a weekly schedule fires on Monday morning.

2

Wistia: Get Account Analytics Timeseries pulls last week's plays, watch time and engagement in one request for the whole window, with the end date set one day past the last day you want, because the analytics end date is exclusive.

3

Wistia: Get Account Top Content returns the week's leading videos in a single call for exactly that window. Do not assemble it from seven daily calls: the response is ranked and truncated, so per day leaderboards cannot be added together without under counting everything below the cut.

4

Wistia: Get Media Analytics fills in the detail for the handful of videos the report is about, and Wistia: Get Media Embed Locations shows where the plays came from, with the page size raised past its small default.

5

The numbers land in a Mattering database next to the previous week's, and a condition compares them: engagement and play rate are recombined by weighting them with the plays and loads underneath rather than by averaging the percentages.

6

A task opens for the marketing owner only when a video moved further than the threshold, with the figures and the embed locations attached. A quiet week produces a database row and no interruption.

Wistia MCP questions, answered

What Wistia operations are supported?

51 nodes on Wistia's modern API at the 2026-07 release: 34 read nodes and 17 action nodes. Account, account usage, search and background job status. Media (list, get, update, delete, copy, import from a URL, archive, restore, move, and the player customizations get and update). Captions (list, create, delete). Tags (list, create, bulk tag media, delete). Folders (list, get, subfolders, create, update, delete) and the deleted media trash (list, restore). Channels, channel episodes, webinars and webinar registrations as reads. Then two reporting families: the stats counters (account, account by date, folder, media, media by date, engagement, visitors, viewing events) and the analytics surface (account analytics and timeseries, account top content, media analytics and timeseries, embed locations, traffic breakdown, form conversions, webinar analytics). Deliberately not included: Wistia's upload host, which takes a multipart file the canvas has no file to give, so URL import is the ingest path here; webhook subscriptions, since this connector polls; and channel, episode and webinar authoring, which are multi step publishing workflows better done in Wistia.

Does this work with Claude, ChatGPT, or other AI assistants?

Yes. Mattering exposes these nodes over MCP, an open protocol for connecting AI assistants to tools, so any MCP-capable assistant, including Claude and ChatGPT, can connect to your workspace and use them. Matter, the assistant built into the platform, uses the same nodes natively. The read half is 34 of the 51 nodes and is the comfortable surface to hand an assistant: finding media, reading performance, checking captions and embed locations all answer questions without changing anything. Keep the four destructive nodes, and the bulk write nodes that queue work you cannot easily unwind, behind a condition or an approval task.

Wistia publishes its own MCP server. Why use this connector?

It does, and that is worth knowing before you choose. Wistia ships an official MCP server alongside a command line tool. We have not used it, so nothing here characterises what it covers: check Wistia's own documentation for that. What this connector is built for is different work. These 51 nodes run on a schedule or a trigger with nobody present, which is what a Monday morning report or a nightly library sweep needs. They chain with every other connector in one workflow, so a new Wistia media item can create a task, write a database row and post to a channel in the same run. The credential is held once in a shared workspace store rather than in each person's local client config. And the destructive nodes carry flags, so a delete can sit behind an approval. They are also ordinary canvas steps, not only assistant tools.

Why does a stats node fail when my media list works fine?

Almost always the token's permission level. Wistia tokens carry a permission level rather than a scope list, and the levels are not nested the way people expect: every stats and analytics node needs Read detailed stats, which is a separate permission that ordinary read access to media and folders does not imply. So a token that lists your entire library can return a permission error on every performance node, which reads like a broken connector and is not. Create a token with the level the workflow needs, or raise the level on the existing one, and note that a few nodes need more still: search, tags and the webinar reads need read access to all data, and every write node needs read, update and delete access.

Can I add up the top content or embed location numbers across weeks?

No, and this is the most important thing to know about Wistia's reporting here. Wistia: Get Account Top Content, Wistia: Get Media Embed Locations and Wistia: Get Media Traffic Breakdown are ranked and truncated: they sort the rows and cut them off at a row limit, and the embed locations read defaults to only ten rows. That means the limit does not merely shorten the response, it changes which rows exist. An entity that falls outside the cut in one sub period does not appear in that period at all, so adding four weekly leaderboards together under counts everything in the middle of the pack while producing a total that looks complete. Ask for the window you actually want in one request, raise the page size when you need depth rather than breadth, and treat these as leaderboards rather than as datasets. The per day series nodes, Wistia: Get Account Stats By Date and Wistia: Get Media Stats By Date, are the ones whose counts do add up cleanly.

Why is my analytics range missing the last day?

Because the two reporting surfaces use opposite conventions and the analytics one is exclusive at the end. An analytics request ending on the last day of a month covers up to the start of that day, so the last day is omitted. The older stats by-date endpoints are inclusive. Every affected field says so in its description, and the fix is simply to ask for one day past the last day you want on the analytics nodes. Two related limits: the analytics nodes refuse a range longer than two years, and Wistia documents the viewing events surface as retaining two years of data.

Can a workflow start when a new video is uploaded?

Not from this connector, which polls rather than listens. There are no trigger nodes here because Wistia's webhooks are a separate surface with their own subscription model that this connector does not manage. For most video workflows a schedule plus Wistia: List Media sorted by creation is enough, since publishing a video is rarely a second sensitive event. If you do need immediate reaction, point a Wistia webhook at a Mattering webhook trigger and use these nodes for everything after that first hop. Note that the ingest node, Wistia: Import Media From URL, is asynchronous in the same way as the other bulk writes: it hands back a job, and Wistia: Get Background Job Status reports what actually happened.

What are Wistia's rate limits?

Wistia documents 600 requests per minute per account, counted across its API hosts together, and answers a request over the limit with a rate limited status and a header saying how long to wait. One node has a much tighter limit of its own that is easy to hit by accident: Wistia documents the media move endpoint at ten requests per five minutes, separate from and far below the account limit, so a loop that moves one video per iteration will stop within a minute. Move media in batches instead. Bulk write endpoints also cap the number of items per request, and each node's description carries its own ceiling.

What happens when a video is deleted?

The honest answer is that Wistia's own documentation is not consistent here, so the connector states both readings rather than picking one. The delete reference describes the media as permanently deleted and not recoverable, while the deleted media reference describes deleted media as restorable for a period and ships a restore endpoint for it, documenting a shorter window on free plans than on paid ones. Do not rely on either reading: check Wistia: List Deleted Media on your own account before you build a workflow that assumes recovery is possible. What is not in dispute is the immediate blast radius, which is why Wistia: Delete Media is flagged: the captions, customizations and stats history go with it, and every embed of that video on every page stops working right away.

51 nodes waiting on the canvas

Put Wistia to work inside your workspace

Connect once, then use every node in visual automations and AI agent workflows across the Mattering platform.

All product names, logos, and brands are property of their respective owners and are used for identification purposes only. Use of these names, logos, and brands does not imply endorsement or affiliation. Wistia is a product of its respective owner; Mattering is an independent platform.

Start Free Explore our Products