GraphQL query examples
Please note that this page is a work in progress
To learn how to use the queries below, read about querying the database through Check API
This is a list of common queries performed on Check API and its responses.
For the collections (so, everytime you ask for a field that contains edges), you must specify how many items should be returned, by passing first or last. For pagination, you can also pass after, which takes the GraphQL ID of a node. For example:
project_medias(first: 10, after: "haghx6xsagh\n") {
edges {
node {
(...)
}
}
}
Table of Contents
- Get team information
- Get team tags
- Get list of status of a team
- Create a project in a team
- Add new auto task to a team
- Get tasks created automatically for project items
- Create an item
- Update an item
- Get data from items
- Get the tasks and responses from an item
- Add a comment on an item
- Get comments of an item
- Get status details of an item
- Bulk update status on items
- Log details of items
- Get report content of an item
- Update tags on items
- Add related item
- List teams a user is related to
- Get medias, their tipline requests and fact-checks
- Get medias, their tipline requests and when they were requested x responded
- Get imported fact-checks
- Create a fact-check
- Update a fact-check
- Delete a fact-check
- Update a claim
- Send item to trash
- Create an explainer
Get the teams that are associated with the token used in authentication and for each team returns the name, slug id and dbid.
Query
query {
me {
teams {
edges {
node {
name
slug
dbid
id
}
}
}
}
}Response
{
"data": {
"me": {
"teams": {
"edges": [
{
"node": {
"name": "Studio Ghibli",
"slug": "studio-ghibli",
"dbid": 3,
"id": "VGVhbS8z\n"
}
},
]
}
}
}
}Get the list of tags present on a team.
Query
query {
team(slug: "studio-ghibli") {
tag_texts {
edges {
node {
text
}
}
}
}
}Response
{
"data": {
"team": {
"tag_texts": {
"edges": [
{
"node": {
"text": "character"
}
},
{
"node": {
"text": "movie"
}
},
{
"node": {
"text": "music"
}
}
]
}
}
}
}In the example below we get all the possible status of a team. We also pass verified to items_count and published_reports_count, it returns the numbers of items and published reports with the status verified. If you don't care about the numbers of a specific status, you can just pass verification_statuses(), without items_count and published_reports_count.
Query
query {
team(slug: "studio-ghibli") {
verification_statuses(items_count_for_status: "verified",
published_reports_count_for_status: "verified")
}
}Response
{
"data": {
"team": {
"verification_statuses": {
"active": "in_progress",
"default": "undetermined",
"label": "Status",
"statuses": [
{
"id": "in_progress",
"label": "Em Progresso",
"locales": {
"en": {
"description": "",
"label": "In Progress",
"message": "We started looking into that!"
},
"fr": {
"description": "",
"label": "en cours"
},
"pt": {
"description": "",
"label": "Em Progresso",
"message": "Começamos a investigar isso!"
}
},
"should_send_message": true,
"style": {
"color": "#ff864d"
}
},
{
"id": "verified",
"label": "Verdadeiro",
"locales": {
"en": {
"description": "",
"label": "True"
},
"fr": {
"description": "",
"label": "Vrai"
},
"pt": {
"description": "",
"label": "Verdadeiro",
"message": ""
}
},
"should_send_message": false,
"style": {
"color": "#41b070"
}
},
{
"id": "misleading",
"label": "Enganoso",
"locales": {
"en": {
"description": "",
"label": "Misleading"
},
"fr": {
"description": "",
"label": "Fallacieux"
},
"pt": {
"description": "",
"label": "Enganoso",
"message": ""
}
},
"style": {
"backgroundColor": "#fc52ff",
"borderColor": "#fc52ff",
"color": "#fc52ff"
},
"published_reports_count": 75,
"items_count": 219
}
]
}
}
}
}Using the team dbid from the first query as id, you can create a project under a team.
Query
mutation {
createProject(input: {
clientMutationId: "1",
title: "The Cat Returns",
team_id: 3
}) {
project {
dbid
}
}
}Response
{
"data": {
"createProject": {
"project": {
"dbid": 12
}
}
}
}Using the team id from the first query, the query below adds a new auto task to a team and gets the updated list of auto tasks as response. The type inside add_auto_task can be geolocation, datetime, free_text, single_choice, multiple_choice or image_upload. For free_text and single_choices, an additional field options should exist inside add_auto_task, which is a JSON string with format [{\"label\":\"Option A\"},{\"label\":\"Option B\"}].
Query
mutation {
updateTeam(input: {
clientMutationId: "1",
id: "VGVhbS8z\n",
add_auto_task: {
label: "Did you watch this movie?",
type: "single_choice",
description: "",
options: [{label: "Yes"},{label: "No"}],
fieldset: "tasks"
}
}) {
team {
team_tasks {
edges {
node {
label
description
options
type
}
}
}
}
}
}Response
{
"data": {
"updateTeam": {
"team": {
"team_tasks": {
"edges": [
{
"node": {
"label": "What's the name of the movie?",
"description": null,
"options": [],
"type": "free_text"
}
},
{
"node": {
"label": "Did you watch this movie?",
"description": "",
"options": [
{
"label": "Yes"
},
{
"label": "No"
}
],
"type": "single_choice"
}
}
]
}
}
}
}
}The metadata/annotations created automatically when creating items are also called "team tasks". This query returns the tasks that will be created automatically when creating items in a team.
The fieldset shows the type "metadata" or "tasks".
Query
query {
team {
team_tasks {
edges {
node {
label
fieldset
description
options
type
required
}
}
}
}
}Response
{
"data": {
"team": {
"team_task": {
"edges": [
{
"node": {
"label": "What's the name of the movie?",
"fieldset": "tasks",
"description": null,
"options": [],
"type": "free_text",
"required": true,
"json_schema": null
}
},
{
"node": {
"label": "Did you watch this movie?",
"fieldset": "metadata",
"description": "",
"options": [
{
"label": "Yes"
},
{
"label": "No"
}
],
"type": "single_choice",
"required": false,
"json_schema": null
}
}
]
}
}
}You can add a link, a claim, a picture or a video to an existent project, optionally with responses to tasks and metadata fields. When answering them, you should send the slug of the task or metadata field in order to identify it.
Query
mutation create {
createProjectMedia(input: {
url: "https://en.wikipedia.org/wiki/Spirited_Away",
clientMutationId: "1"
}) {
project_media {
title
dbid
id
}
}
}Response
{
"data": {
"createProjectMedia": {
"project_media": {
"title": "Spirited Away - Wikipedia",
"dbid": 108,
"id": "UHJvamVjdE1lZGlhLzEwOA==\n"
}
}
}
}Similar to the above.
Query
mutation create {
createProjectMedia(input: {
quote: "Once you've met someone you never really forget them. It just takes a while for your memories to return.",
clientMutationId: "1"
}) {
project_media {
title
dbid
id
}
}
}
Response
{
"data": {
"createProjectMedia": {
"project_media": {
"title": "Once you've met someone you never really forget them. It just takes a while for your memories to return.",
"dbid": 107,
"id": "UHJvamVjdE1lZGlhLzEwNw==\n"
}
}
}
}import requests
import json
text = "Text for the item 🎉"
mutation="""mutation create {
createProjectMedia(input: {
quote: """ + json.dumps(text, ensure_ascii=False) + """
}) {
project_media {
id
dbid
}
}
}"""
req = requests.request("POST",
"https://check-api.checkmedia.org/api/graphql",
headers={
"X-Check-Token": "<TOKEN>",
"X-Check-Team": "<WORKSPACE_SLUG>",
},
data = {"query": mutation}
)
print(req.text)This is similar to create a link above, but adding set_tags and set_status to the input. You can also add these fields when you create a claim to create a claim with tags and a status.
-
set_tagsexpects an array of strings -
set_statusexpects a string that matches theidof one of your statuses. By default these are one ofundetermined,in_progress,not_applicable,false, andverified, but you may have custom statuses defined in your workspace. To learn their IDs you can follow these instructions
Query
mutation create {
createProjectMedia(input: {
url: "https://en.wikipedia.org/wiki/Spirited_Away",
clientMutationId: "1",
set_tags: ["movie", "animation"], # Array of strings (optional)
set_status: "in_progress", # One of the statuses supported by the workspace (the identifier, not the label) (optional)
}) {
project_media {
title
dbid
id
tags {
edges {
node {
id
tag_text
}
}
}
status
}
}
}Response
{
"data": {
"createProjectMedia": {
"project_media": {
"title": "Spirited Away - Wikipedia",
"dbid": 108,
"id": "UHJvamVjdE1lZGlhLzEwOA==\n"
"tags": {
"edges": [
{
"node": {
"id": "VGFnLzI5NDA=\n",
"tag_text": "movie"
}
},
{
"node": {
"id": "VGFnLzI5Mzk=\n",
"tag_text": "animation"
}
}
]
},
"status": "in_progress"
}
}
}
}The example below creates an item and answers some team tasks.
The format to send the tasks response is set_tasks_responses = { task_slug (string) => response (string) }, where:
-
task_slugis the slug of a task label. Example:What's the name of the moviewill bewhat_s_the_name_of_the_movie -
responseis the task answer To see the available team tasks and labels, see how to get metadata/annotations created automatically for items
Query
mutation create {
createProjectMedia(input: {
url: "https://en.wikipedia.org/wiki/Spirited_Away",
clientMutationId: "1",
set_tasks_responses: "{ \"what_s_the_name_of_the_movie\": \"Spirited Away\", \"did_you_watch_this_movie\": \"Yes\" }"
}) {
project_media {
title
dbid
id
}
}
}Response
{
"data": {
"createProjectMedia": {
"project_media": {
"title": "Spirited Away - Wikipedia",
"dbid": 108,
"id": "UHJvamVjdE1lZGlhLzEwOA==\n"
}
}
}
}We can’t upload files through our graphiQL client, but you can do that by using a tool like postman, curl, among other options.
Here we have an example of how to create an item with an uploaded image and one with video using curl.
curl \
-H 'X-Check-Token: <your_token>' \
-H 'X-Check-Team: <your_check_workspace_slug>' \
-F query='mutation create { createProjectMedia(input: { media_type: "UploadedImage" }) { project_media { id dbid } } }' \
-F file=@/path/your_file.png \
"https://check-api.checkmedia.org/api/graphql"
curl \
-H 'X-Check-Token: <your_token>' \
-H 'X-Check-Team: <your_check_workspace_slug>' \
-F query='mutation create { createProjectMedia(input: { media_type: "UploadedVideo" }) { project_media { id dbid } } }' \
-F file=@/path/your_file.mp4 \
"https://check-api.checkmedia.org/api/graphql"
import requests
media=open("chair-20-sd-bar.mp4","rb").read()
mutation="""mutation create {
createProjectMedia(input: {
media_type: "UploadedVideo"
}) {
project_media {
id
dbid
}
}
}"""
req = requests.request("POST",
"https://check-api.checkmedia.org/api/graphql",
headers={
"X-Check-Token": "<TOKEN>",
"X-Check-Team": "<WORKSPACE_SLUG>",
},
data = {"query": mutation},
files = {"file": ("chair-20-sd-bar.mp4", media, "video/mp4")}
)
print(req.text)It's possible to send or restore a single trashed item using the mutation to update an item.
An item is in the trash when the attribute archived is true. If you want to send the items to trash, the mutation must include archived: 1. If the goal is to restore the items from trash, must include archived: 0.
Query
mutation {
updateProjectMedia(input: {
clientMutationId: "1",
id: "UHJvamVjdE1lZGlhLzEwOA==\n",
archived: 1
}) { affectedId }
}
Response
{
"data": {
"updateProjectMedia": {
"affectedId": "UHJvamVjdE1lZGlhLzEwOA==\n"
}
}
}It's also possible to bulk send or restore trashed items using the mutation to update an item.
An item is in the trash when the attribute "archived" is true. If you want to send the items to trash, the mutation must include "archived": 1. If the goal is to restore the items from trash, must include "archived": 0.
Query
mutation {
updateProjectMedias(input: {
clientMutationId: "1",
ids: [ "UHJvamVjdE1lZGlhLzI0MDI=\n", "UHJvamVjdE1lZGlhLzI0MDE=\n", "UHJvamVjdE1lZGlhLzI0MDQ=\n"],
action: "archived", params: "{\"archived\": 1}"
}) { ids, team { dbid } }
}
Response
{
"data": {
"updateProjectMedias": {
"ids": [
"UHJvamVjdE1lZGlhLzI0MDI=\n",
"UHJvamVjdE1lZGlhLzI0MDE=\n",
"UHJvamVjdE1lZGlhLzI0MDQ=\n"
],
"team": {
"dbid": 3
}
}
}
}To customize the title and description displayed on analysis sidebar and report you'll need the id of the project_media's last_status_obj. Check Get status details of an item to know how to get the id.
The title and description are saved as annotation fields. Please note the set_fields field should be sent in JSON string format.
The example below shows how to add a customized title and description on item that has the status with the id sent on input.
Query
mutation {
updateDynamic(input: {
id: "RHluYW1pYy83MDE1NTY1\n",
set_fields: "{ \"title\": \"Spirited Away\", \"content\": \"A 2001 Japanese animated fantasy film\" }"
clientMutationId: "1"
}) {
project_media {
title
description
}
}
}Response
{
"data": {
"updateDynamic": {
"project_media": {
"description": "A 2001 Japanese animated fantasy film",
"title": "Spirited Away"
}
}
}
}To update the verification status of an item, e.g: from in progress to verified. You will need to update the Dynamic Annotation. To do that, you will need it's id. Then you will be able to use that id to update the verification status. In the example below, "147" is the item ID, while "52" is the team ID. "0" is hard-coded for legacy reasons - it used to be the folder ID.
Query
query {
project_media(ids: "147,0,52"){
dynamic_annotation_verification_status {
id
}
}
}Response
{
"data": {
"project_media": {
"dynamic_annotation_verification_status": {
"id": "RHluYW1pYy80ODY=\n"
}
}
}
}Query
mutation {
updateDynamic(input: {
id: "RHluYW1pYy80ODY=\n"
set_fields: "{\"verification_status_status\":\"verified\"}"
}) {
project_media {
dbid,
id,
last_status
}
}
}Response
{
"data": {
"updateDynamic": {
"project_media": {
"id": "UHJvamVjdE1lZGlhLzE0Nw==\n",
"dbid": 147,
"last_status": "verified"
}
}
}
}With the slug or the dbid of a team you can get the projects and the data from project medias.
Query
query {
team(id: 3) {
name
projects_count
projects(last: 3) {
edges {
node {
title
project_medias(first: 1) {
edges {
node {
type
title
picture
}
}
}
}
}
}
}
}Response
{
"data": {
"team": {
"name": "Studio Ghibli",
"projects_count": 6,
"projects": {
"edges": [
{
"node": {
"title": "Spirited Away",
"project_medias": {
"edges": [
{
"node": {
"type": "Link",
"title": "Slackline-Tutorial: How to take nice Leashfalls on Highlines",
"picture": "https://i.ytimg.com/vi/kQsrpDnOuKU/maxresdefault.jpg"
}
}
]
}
}
},
{
"node": {
"title": "The Cat Returns",
"project_medias": {
"edges": [
{
"node": {
"type": "Link",
"title": "The Cat Returns - Wikipedia",
"picture": "https://upload.wikimedia.org/wikipedia/en/8/8e/Cat_Returns.jpg"
}
}
]
}
}
},
{
"node": {
"title": "The Tale of the Princess Kaguya",
"project_medias": {
"edges": [
{
"node": {
"type": "Link",
"title": "The Tale of the Princess Kaguya - Wikipedia",
"picture": "https://upload.wikimedia.org/wikipedia/en/6/68/The_Tale_of_the_Princess_Kaguya_%28poster%29.jpg"
}
}
]
}
}
}
]
}
}
}
}Sending the dbid of a project as id you can get the data from project medias. The query below is also returning the field archived, that has a boolean value and is true when the item is on trash and false otherwise
Query
query {
project(id: "4") {
title
project_medias(first: 2) {
edges {
node {
archived
title
media {
picture
url
}
}
}
}
}
}Response
{
"data": {
"project": {
"title": "Spirited Away",
"project_medias": {
"edges": [
{
"node": {
"archived": true,
"title": "Spirited Away",
"media": {
"picture": "https://upload.wikimedia.org/wikipedia/en/d/db/Spirited_Away_Japanese_poster.png",
"url": "https://en.wikipedia.org/wiki/Spirited_Away"
}
}
},
{
"node": {
"archived": false,
"title": "Once you've met someone you never really forget them. It just takes a while for your memories to return.",
"media": {
"picture": "",
"url": null
}
}
}
]
}
}
}
}Sending the dbid of an item as ids you can request the tasks and the responses.
Query
query {
project_media(ids: "108") {
tasks {
edges {
node {
label
first_response_value
}
}
}
}
}Response
{
"data": {
"project_media": {
"tasks": {
"edges": [
{
"node": {
"label": "Do you recommend this movie?",
"first_response_value": null
}
},
{
"node": {
"label": "What's the name of the movie?",
"first_response_value": "Spirited Away"
}
},
{
"node": {
"label": "Did you watch this movie?",
"first_response_value": "Yes"
}
}
]
}
}
}
}A comment can be added to an item sending the type ProjectMedia as annotated_type and the dbid as annotated_id
Query
mutation {
createComment(input: {
clientMutationId: "1",
text: "A comment created on an item",
annotated_id: "108",
annotated_type: "ProjectMedia"
}) {
comment {
text
}
}
}Response
{
"data": {
"createComment": {
"comment": {
"text": "A comment created on an item"
}
}
}
}The comments of an item are saved as annotations and displayed in Notes tab. If you want to display the comments of an specific item, you need to send the argument ids with the dbid (database id) of the item.
The example below shows how to query the item with dbid 123456 and get the comments and some details: the text added, the creation time as integer and the name of the user who added the comment.
Query
query {
project_media(ids: "123456") {
annotations(annotation_type: "comment") {
edges {
node {
... on Comment {
text
created_at
annotator {
user {
name
}
}
}
}
}
}
}
}Response
{
"data": {
"project_media": {
"annotations": {
"edges": [
{
"node": {
"annotator": {
"user": {
"name": "Totoro"
}
},
"created_at": "1596166712",
"text": "A comment created on an item"
}
},
{
"node": {
"annotator": {
"user": {
"name": "Chihiro"
}
},
"created_at": "1596166481",
"text": "This is a very nice movie!"
}
}
]
}
}
}
}The status of an item is a dynamic annotation. This annotation has fields for the verification status and for custom title and description. To display the last status details of an specific item, you need to send the argument ids with the dbid (database id) of the item.
The example below shows how to query the item with dbid 108 and get the last status id, data and content and get the title and description of the item.
Query
query {
project_media(ids: "108") {
last_status_obj {
id
data
content
}
title
description
}
}Response
{
"data": {
"project_media": {
"last_status_obj": {
"content": "[{\"id\":1752,\"annotation_id\":1666,\"field_name\":\"verification_status_status\",\"annotation_type\":\"verification_status\",\"field_type\":\"select\",\"value\":\"in_progress\",\"value_json\":{},\"created_at\":\"2019-03-10T12:12:55.593Z\",\"updated_at\":\"2019-03-10T12:12:55.593Z\",\"formatted_value\":\"In progress\"}]",
"data": {
"fields": [
{
"annotation_id": 1666,
"annotation_type": "verification_status",
"created_at": "2019-03-10T12:12:55.593Z",
"field_name": "verification_status_status",
"field_type": "select",
"formatted_value": "In progress",
"id": 6939752,
"updated_at": "2019-03-10T12:12:55.593Z",
"value": "in_progress",
"value_json": {}
}
],
"indexable": "in_progress"
},
"id": "RHluYW1pYy83MDE1NTY1\n"
},
"title": "Spirited Away - Wikipedia"
"description": "A Japanese animated fantasy film"
}
}
}Statuses are bulk updated on items using the updateProjectMedias interface. You must provide an action of a string "update_status" and an ids field that is an array of the id of each individual item you are updating (it's a long string, not the dbid which is a number). Then under params you send a JSON string of an object containing a status property whose value is a string corresponding to a status id. You can find these ids by following instructions here and looking for id under each object in the statuses array.
In the example below we are setting the status "verified" on two items.
Query
mutation {
updateProjectMedias(input:{
"action": "update_status",
"clientMutationId": "1",
"ids": [
"UHJvamVjdE1lZGlhLzIxNA==\n", # the individual `id` for each item you wish to remove tags from (not the `dbid`)
"UHJvamVjdE1lZGlhLzIxNQ==\n"
],
"params": "{\"status\":\"verified\"}" # a JSON string of an object containing a `status` property whose value is one of the statuses supported by the workspace (the identifier, not the label)
}) {
updated_objects {
dbid
id
status
}
}
}Response
{
"data": {
"updateProjectMedias": {
"updated_objects": [
{
"dbid": 218,
"id": "UHJvamVjdE1lZGlhLzIxOA==\n",
"status": "verified",
},
{
"dbid": 219,
"id": "UHJvamVjdE1lZGlhLzIxOQ==\n",
"status": "verified",
}
],
"clientMutationId": "1"
}
}
}On Check the log details are displayed on Timeline. When getting the data you can see the changes on field object_changes_json in JSON format.
Query
query me {
project_media(ids: "108") {
title
log(last: 4) {
edges {
node {
event_type
object_changes_json
created_at
}
}
}
}
}Response
{
"data": {
"project_media": {
"title": "Spirited Away",
"log": {
"edges": [
{
"node": {
"event_type": "update_dynamicannotationfield",
"object_changes_json": "{\"value\":[\"--- undetermined\\n\",\"--- verified\\n\"]}",
"created_at": "1591989930"
}
},
{
"node": {
"event_type": "update_projectmedia",
"object_changes_json": "{\"archived\":[false,true]}",
"created_at": "1591994647"
}
},
{
"node": {
"event_type": "update_projectmedia",
"object_changes_json": "{\"archived\":[true,false]}",
"created_at": "1591994829"
}
},
{
"node": {
"event_type": "update_dynamicannotationfield",
"object_changes_json": "{\"value\":[\"--- '{\\\"title\\\":\\\"Spirited Away\\\",\\\"description\\\":\\\"A\\\"}'\\n\",\"--- '{\\\"title\\\":\\\"Spirited Away\\\",\\\"description\\\":\\\"A 2001 Japanese animated fantasy film\\n written and directed by Hayao Miyazaki\\\"}'\\n\"],\"value_json\":[\"{\\\"title\\\":\\\"Spirited Away\\\",\\\"description\\\":\\\"A\\\"}\",\"{\\\"title\\\":\\\"Spirited Away\\\",\\\"description\\\":\\\"A 2001 Japanese animated fantasy film written and directed by Hayao Miyazaki\\\"}\"]}",
"created_at": "1591995024"
}
}
]
}
}
}
}The report of an item is a dynamic annotation with annotation_type: report_design.
The example below shows how to query the item with dbid 110 and get the report id, dbid and data and get the title and description of the item.
Query
query {
project_media(ids: "110") {
title
description
annotation(annotation_type: "report_design") {
id
dbid
data
}
}
}Response
{
"data": {
"project_media": {
"annotation": {
"data": {
"first_published": "1621627340",
"last_published": "1621627366",
"options": [
{
"description": "This summer's Olympics will go ahead even if Tokyo remains in a state of a emergency, says IOC vice-president John Coates.",
"headline": "State of emergency won't stop Games - IOC",
"image": "https://assets.checkmedia.org/medias/f98cee0dfe2f8e7acc12d258ec478d93/picture.jpg",
"introduction": "Regarding what you sent to us on {{query_date}}, we reached the conclusion that it is *{{status}}*.",
"language": "en",
"previous_published_status_label": "True",
"status_label": "True",
"text": "",
"theme_color": "#41b070",
"title": "",
"url": "meedan.com",
"use_introduction": true,
"use_text_message": false,
"use_visual_card": true,
"visual_card_url": "https://check-api-live.s3.eu-west-1.amazonaws.com/narcissus/screenshot-bf9fa222789763ff732e1724545e6f017c3357cd.png"
}
],
"published_count": 1,
"state": "published"
},
"dbid": "9161120",
"id": "QW5ub3RhdGlvbi85MTYxMTIw\n"
},
"description": "This summer's Olympics will go ahead even if Tokyo remains in a state of a emergency, says IOC vice-president John Coates.",
"title": "State of emergency won't stop Games - IOC"
}
}
}A tag can be added to an item sending the type ProjectMedia as annotated_type and the dbid as annotated_id
Query
mutation {
createTag(input:{
clientMutationId: "1",
tag: "character",
annotated_id: "27",
annotated_type: "ProjectMedia"
}) {
tagEdge {
node {
dbid
tag
}
}
}
}Response
{
"data": {
"createTag": {
"tagEdge": {
"node": {
"dbid": "740",
"tag": "1"
}
}
}
}
}Multiple tags can be added to multiple items by using createTags() and sending an array containing ProjectMedia as annotated_type and the dbid as annotated_id along with the tag text itself as tag. In the example below we add apple and orange to one item and orange to another item.
Query
mutation {
createTags(input:{
clientMutationId: "1",
inputs: [
{
tag: "apple",
annotated_id: "27",
annotated_type: "ProjectMedia"
},
{
tag: "orange",
annotated_id: "27",
annotated_type: "ProjectMedia"
},
{
tag: "orange",
annotated_id: "28",
annotated_type: "ProjectMedia"
},
]
}) {
team {
id
}
}
}Response
Note: bulk updates don't give specific responses so here we are just asking for the team
idas confirmation in response. You could query for the full list of team tags or whatever else you like here.
{
"data": {
"createTags": {
"team": {
"id": "id": "VGVhbS81\n"
}
}
}
}Tags are bulk removed from items using the updateProjectMedias interface. You must provide an action of a string "remove_tags" and an ids field that is an array of the id of each individual item you are updating (it's a long string, not the dbid which is a number). Then under params you send a JSON string of an object containing a tags_text property whose value is a comma-separated string of numbers. these numbers correspond to the tag field of an individual tag (rather than tag_text as you might expect). You can find the tag value for each tag by doing a query for all team tags but adding tag to the list of properties you want sent back to you (alongside text in that linked example).
In the example below we are removing tags 5 and 6 from two items.
Query
mutation {
updateProjectMedias(input:{
"action": "remove_tags",
"clientMutationId": "1",
"ids": [
"UHJvamVjdE1lZGlhLzIxNA==\n", # the individual `id` for each item you wish to remove tags from (not the `dbid`)
"UHJvamVjdE1lZGlhLzIxNQ==\n"
],
"params": "{\"tags_text\":\"5,6\"}" # a JSON string of an object containing a `tags_text` property whose value is a comma-separated string of numbers. these numbers correspond to the `tag` field of an individual tag (rather than `tag_text`)
}) {
updated_objects {
dbid
id
tags {
edges {
node {
tag
tag_text
}
}
}
}
}
}Response
{
"data": {
"updateProjectMedias": {
"updated_objects": [
{
"dbid": 214,
"id": "UHJvamVjdE1lZGlhLzIxNA==\n",
"tags": {
"edges": [
{
"node": {
"tag": "1",
"tag_text": "apple"
}
},
{
"node": {
"tag": "2",
"tag_text": "orange"
}
}
]
}
},
{
"dbid": 215,
"id": "UHJvamVjdE1lZGlhLzIxNQ==\n",
"tags": {
"edges": [
{
"node": {
"tag": "1",
"tag_text": "apple"
}
}
]
}
}
],
"clientMutationId": "1"
}
}
}Items can be listed as secondary items of other items by adding them as related. In order to add a related item, you'll need the id of the child item and the dbid of the parent item, that should be sent as related_to_id.
Query
mutation {
updateProjectMedia(input:{
clientMutationId: "1",
id:"UHJvamVjdE1lZGlhLzEwNw==\n",
related_to_id: 26
}) {
related_to {
dbid
title
}
}
}Response
{
"data": {
"updateProjectMedia": {
"related_to": {
"dbid": 26,
"title": "Zeniba"
}
}
}
}A team_user is a relation between a Team and a User on Check. On the example below, we get the list of team a user is related to and some fields: the id (used to update the relation), the team's name, the role and status the user has on the team.
Query
query {
me {
team_users {
edges {
node {
id
team {
name
}
role
status
}
}
}
}
}Response
{
"data": {
"me": {
"team_users": {
"edges": [
{
"node": {
"id": "VGVhbVVzZXIvNjQ=\n",
"role": "journalist",
"status": "banned",
"team": {
"name": "Aardman Animations"
}
}
},
{
"node": {
"id": "VGVhbVVzZXIvMTA4OA==\n",
"role": "contributor",
"status": "member",
"team": {
"name": "Laika"
}
}
},
{
"node": {
"id": "VGVhbVVzZXIvOTU4\n",
"role": "owner",
"status": "member",
"team": {
"name": "Blue Sky Studios"
}
}
},
{
"node": {
"id": "VGVhbVVzZXIvNDMzMg==\n",
"role": "editor",
"status": "member",
"team": {
"name": "Studio Ghibli"
}
}
}
]
}
}
}
}Query
query {
search(query: "{\"channels\":[\"any_tipline\"],\"eslimit\":50,\"esoffset\":0}") { # Change eslimit and esoffset in order to paginate
number_of_results
medias {
edges {
node {
id
last_status
requests { # Tipline requests
edges {
node {
id
value_json
}
}
}
media { # Media attached
id
type
quote # Present only if type is claim (text)
url # Present only if type is link
file_path # Present only if type is an uploaded audio, image or video
}
claim_description {
description # The claim being fact-checked
fact_check { # The fact-check
id
title
summary
url
}
}
}
}
}
}
}Query
query {
search(query: "{\"eslimit\":10000,\"range\":{\"created_at\":{\"start_time\":\"2024-09-01T00:00:00.000Z\",\"end_time\":\"2024-10-01T00:00:00.000Z\"}}}") {
medias(first: 10000) {
edges {
node {
full_url
requests(first: 10000, includeChildren: true) {
edges {
node {
dbid
created_at
responded_at
}
}
}
}
}
}
}
}Query
query {
search(query: "{\"esoffset\":0,\"eslimit\":100,\"channels\":[\"1\"]}") {
number_of_results
medias {
edges {
node {
id
claim_description {
fact_check {
id
url
}
}
}
}
}
}
}Create a fact-check not related to any media, but with a claim, tags, status and the fact-check itself.
Query
mutation create {
createProjectMedia(input: {
media_type: "Blank", # To be used when there is no media... if there is a media, replace this by either "url" (for link media) or "quote" (for text media)... when "url" is present, creation takes longer because it's going to be parsed
channel: { main: 1 }, # This will make this item be listed under "Imported reports"
set_tags: ["science", "astronomy"], # Array of strings (optional)
set_status: "verified", # One of the statuses supported by the workspace (the identifier, not the label) (optional)
set_claim_description: "Earth is round.",
set_fact_check: {
title: "TRUE: The Earth is round",
summary: "Scientific evidences show that the Earth is round.",
url: "https://en.wikipedia.org/wiki/Earth",
language: "en", # Must be the ISO-361 code of a language supported by the workspace (optional)
publish_report: true # If `true`, it can match incoming tipline queries (optional)
}
}) {
project_media {
id # Use this ID to update the item
full_url # Link to the created item
claim_description {
fact_check {
id # Use this ID to update the fact-check
}
}
}
}
}Response
{
"data": {
"createProjectMedia": {
"project_media": {
"id": "The item Base 64 ID",
"full_url": "https://checkmedia.org/slug/project/1234/media/1234",
"claim_description": {
"id": "The claim Base 64 ID"
"fact_check": {
"id": "The fact-check Base 64 ID"
}
}
}
}
}
}Example with cURL
#!/bin/bash
export CHECK_TOKEN="your-api-key"
export CHECK_WORKSPACE_SLUG="your-check-workspace-slug"
query=$(cat <<-END
{
"query" : "mutation create {
createProjectMedia(input: {
media_type: \"Blank\",
channel: { main: 1 },
set_tags: [\"science\", \"astronomy\"],
set_status: \"verified\",
set_claim_description: \"Earth is round.\",
set_fact_check: {
title: \"TRUE: Earth is round\",
summary: \"Scientific evidences show that Earth is round.\",
url: \"https://en.wikipedia.org/wiki/Earth\",
language: \"en\",
publish_report: true
}
}) {
project_media {
id # Keep track of the item ID so you can update or delete it later
claim_description {
id # Keep track of the claim ID so you can update or delete it later
fact_check {
id # Keep track of the fact-check ID so you can update or delete it later
}
}
}
}
}"
}
END
)
echo $query > query.json
curl -X POST \
-H "X-Check-Token: $CHECK_TOKEN" \
-H "Content-Type: application/json" \
-H "X-Check-Team: $CHECK_WORKSPACE_SLUG" \
-d @query.json \
"https://check-api.checkmedia.org/api/graphql"
echoHere is how each field maps to what is displayed in the Check UI:
When creating a fact-check that includes a text or claim, use the `quote` field to specify the text content. This is useful for highlighting specific claims or statements being fact-checked.
Query
mutation create {
createProjectMedia(input: {
media_type: "Claim", # Specifies that the attached media is a text (claim)
quote: "The specific claim or statement you are fact-checking.",
channel: { main: 1 },
set_tags: ["tag1", "tag2"],
set_status: "verified",
set_claim_description: "Description of the claim.",
set_fact_check: {
title: "Fact-check title",
summary: "Summary of the fact-check findings.",
url: "URL to the full fact-check report.",
language: "en",
publish_report: true
}
}) {
project_media {
id
full_url
}
}
}To add a fact-check that includes a link, perhaps to an external source or article, use the url field.
Query
Copy code
mutation create {
createProjectMedia(input: {
media_type: "Link", # Specifies that the attached media is a link
url: "https://example.com/the-specific-article-or-source-being-fact-checked",
channel: { main: 1 },
set_tags: ["tag1", "tag2"],
set_status: "verified",
set_claim_description: "Description of the claim related to the link.",
set_fact_check: {
title: "Fact-check title for the linked article",
summary: "Summary of the fact-check findings for the linked article.",
url: "URL to the full fact-check report.",
language: "en",
publish_report: true
}
}) {
project_media {
id
full_url
}
}
}When adding a fact-check that includes an image, you'll need to upload the image file. Since file uploads might not be directly supported through GraphQL mutations, you may need to use a multipart/form-data request in a tool that supports file uploads, like Postman or a custom client.
Example with cURL
curl -X POST \
-H "X-Check-Token: your-api-token" \
-H "X-Check-Team: your-check-workspace-slug" \
-F query='mutation create { createProjectMedia(input: { media_type: "UploadedImage", set_status: "verified", set_claim_description: "Description of the claim related to the image.", set_fact_check: { title: "Fact-check title for the image", summary: "Summary of the fact-check findings for the image.", language: "en", publish_report: true } }) { project_media { id dbid } } }' \
-F file=@/path/to/image/file.jpg \
"https://check-api.checkmedia.org/api/graphql"Make sure to replace /path/to/image.jpg with the actual path to the image file you wish to upload, and fill in your API token and workspace slug. This example demonstrates how to send a file along with your GraphQL query in a multipart/form-data request, which is commonly supported by various HTTP clients and tools.
When adding a fact-check, you can use the set_original_claim field to import a URL as the original claim. This URL can be in the following formats:
- Video URL: This will create a video as the original claim.
- Audio URL: This will create an audio as the original claim.
- Image URL: This will create a picture/image as the original claim.
- Normal URL: A Link will be created as the original claim when the URL does not point to any type of media.
- If the value in
set_original_claimis not a URL, then the original claim will be created in plain-text.
Here is an example of how to use this field:
Query
mutation {
createProjectMedia(input: {
set_original_claim: "https://someurl.com/somefile.mp3"
}) {
project_media {
title
type
url
}
}
}This example will create an audio as the original claim.
Query
mutation update {
updateFactCheck(input: {
id: "The fact-check Base 64 ID",
title: "New fact-check title",
summary: "New fact-check summary",
url: "New fact-check URL",
language: "New fact-check language"
}) {
fact_check {
updated_at
}
}
}Response
{
"data": {
"updateFactCheck": {
"fact_check": {
"updated_at": "1679006014"
}
}
}
}The example below deletes a fact-check based on its ID.
Query
mutation {
destroyFactCheck(input: {
id: "The fact-check Base 64 ID"
}) { deletedId }
}Response
{
"data": {
"destroyFactCheck": {
"deletedId": "The fact-check Base 64 ID"
}
}
}Query
mutation update {
updateClaimDescription(input: {
id: "The claim Base 64 ID",
description: "New claim description",
context: "New context for the claim"
}) {
claim_description {
updated_at
}
}
}Response
{
"data": {
"updateClaimDescription": {
"claim_description": {
"updated_at": "1679006014"
}
}
}
}Query
mutation createExplainer {
createExplainer(input: {
title: "New explainer title",
description: "New explainer summary",
url: "https://explainer.url/", # Optional
language: "en", # Must be the ISO-361 code of a language supported by the workspace (optional)
tags: ["tag1", "tag2"] # Optional
}) {
explainer {
id
}
}
}Response
{
"data": {
"createExplainer": {
"explainer": {
"id": "RXhwbGFpbmVyLzEK"
}
}
}
}