Integration · Productivity

Connect Claude & ChatGPT to Check

Ask your AI to find claims in your Check workspace, add a new link or quote, or draft a fact-check. It can search media items, add claims and tags, link related items and update verification status. It can create and update fact-checks and explainers, read tipline requests and reply to the person who sent them, with your approval if you want it.

Connect Check free

Free plan, no credit card. Takes about a minute.

Try asking

“Find items in our workspace about the new bridge toll.”
The AI searches media items by keyword and lists the matching claims, links and images.
“Has anyone already sent us this link?”
It looks up the media items that were submitted with that URL.
“Add this article to Check and tag it health.”
It creates a media item from the URL and adds the tag.

What your AI can do

Media items and claims

Add links and quotes, search items, add claims and tags, link related items and update verification status.

Fact-checks and explainers

Create, read and update fact-checks with rating, summary and URL, and attach explainers to items.

Tipline

Read tipline requests and reply to the people who sent them.

“Write a fact-check for this claim, rated false, with a short summary.”
It creates a fact-check with title, summary and rating for the claim. It is only published to your tipline audience if you ask for that.
“Attach our explainer on vaccine testing to item 4812.”
It finds the explainer and attaches it to the media item.
“Reply to tipline request 3310 with a link to our fact-check.”
It reads the request and sends the reply. This is a real message to the person who wrote to your tipline.

22 tools for Check

These tools are switched on when you connect. You can switch any of them off, or require your approval before it runs.

add_claimChanges dataAdd the claim being checked to a media item: `input` with `project_media_id`, the claim `description` and optional `context`.
add_tagChanges dataAdd a tag to an item: `input` with the `tag` text and the item's `annotated_id` and `annotated_type` (e.g. ProjectMedia).
attach_explainer_to_mediaChanges dataAttach an explainer to a media item: `input` with `explainerId` and `projectMediaId`.
create_explainerChanges dataCreate an explainer (background article): `input` with `title`, `description`, optional `url`, `language` and `tags`.
create_fact_checkChanges dataCreate a fact-check: `input` with `title` and `summary`, plus optional `url`, `rating`, `language`, `tags` and the `claim_description_id` it answers. Setting `publish_report` publishes it to the tipline audience.
create_media_itemChanges dataAdd a media item to the workspace: `input` with a `url` or a text `quote`, plus optional `media_type`, `set_title`, `set_tags`, `set_status` and `set_claim_description`.
find_media_by_urlChanges dataFind media items that were submitted with a given `url`, with `first`/`after` paging.
get_current_userChanges dataGet the Check user the API key acts as, including the current workspace. No inputs needed.
get_explainerChanges dataGet one explainer (background article) by `id`.
get_fact_checkChanges dataGet one fact-check by `id`, with title, summary, rating, URL and language.
get_media_itemChanges dataGet one media item with its title, status, claim and fact-check. `ids` is written as "itemId,workspaceId".
get_sourceChanges dataGet one source (the author or account behind media) by `id`.
get_tipline_requestChanges dataGet one tipline request (a message a user sent to your tipline) by `id`.
get_workspaceChanges dataGet the current Check workspace (team), or another one by `id` or `slug`, with its settings and counts.
link_media_itemsChanges dataLink two media items as related or similar: `input` with `source_id`, `target_id` and optional `relationship_type`.
remove_tagDeletesRemove a tag from an item by the tag's `input.id`.
reply_to_tipline_requestVisible to othersReply to a tipline user: `input` with `inReplyToId` (the request ID) and `message`. This sends a real message to that person.
search_media_itemsChanges dataSearch media items (claims, links, images, videos) in the workspace. `query` is a JSON string such as {"keyword":"search words"}.
update_claimChanges dataUpdate a claim by `input.id`: its `description` or `context`.
update_explainerChanges dataUpdate an explainer by `input.id`: title, description, url, language or tags.
update_fact_checkChanges dataUpdate a fact-check by `input.id`: title, summary, url, rating, language or tags.
update_media_itemChanges dataUpdate a media item by `input.id`: `custom_title`, `read`, `archived` (e.g. send to trash) or `related_to_id`.

Set up in three steps

1
Pick the app
Create a free PipMCP account and choose Check from the app list.
2
Paste your key
Paste your Check workspace slug and an API key. You must be the workspace owner: open your workspace Settings > Integrations, enable API Access and click + New API Key. Keys have Editor permissions.
3
Add the link to your AI
You get a personal MCP link. Add it to Claude, ChatGPT or Cursor:
1. Click your name, then Settings › Connectors › Add custom connector.
2. Paste your link as the Remote MCP server URL.
3. Switch it on from the + menu in a chat.

Questions

What can the AI do in Check?

The AI can search and read media items, fact-checks, explainers, sources, tipline requests and your workspace. It can add media items, claims and tags, link related items, update items, claims and verification status, create and update fact-checks and explainers, attach an explainer to an item, remove a tag and reply to a tipline user.

Does the AI see my Check credentials?

No. Your Check API key is encrypted at rest and never shown to the AI. After you save it, it is not shown again, not even to you. The AI only sees the results of the tools it calls.

Can I control what the AI is allowed to do?

Yes. You choose which tools are switched on, so you can start with searching and reading. Replying to a tipline user sends a real message, and removing a tag cannot be undone. Both can require your approval before they run. Every tool call is logged.

Does it work with ChatGPT?

Yes. In ChatGPT go to Settings › Apps & Connectors › Advanced and turn on Developer mode, then add your PipMCP link. Developer mode needs a paid ChatGPT plan: Plus, Pro, Business or Enterprise. The same link also works in Claude (Settings › Connectors › Add custom connector), Cursor and other MCP clients.

What does it cost?

PipMCP has a free plan with no credit card required. Paid plans bill per completed task. You also need your own Check account.

Who can create a Check API key?

Only the workspace owner. In the workspace Settings > Integrations, the owner enables API Access and creates a key. The key has Editor permissions, and the workspace slug you paste tells PipMCP which workspace to work in.