Before you start, make sure you have a team setup on Check and an API key. Ask the Check team for an API key for your team if you don't have one.
You can make requests either using a command line tool, like cURL, or use our API Explorer: https://check-api.checkmedia.org/graphiql.
If you want to use our API Explorer, you just need to paste the GraphQL query on the left and press the "play" button on top.
If you prefer to use a command line tool, the format is: curl -XPOST -H 'X-Check-Token: <API Key>' -d "query=<query>" https://check-api.checkmedia.org/api/graphql.
The available queries are:
Query
query {
me {
current_team {
id
dbid
}
}
}
Response
{
"data": {
"me": {
"current_team": {
"id": "VGVhbS81MjY=\n",
"dbid": 526
}
}
}
}
List all projects under the current team but ask only for the title field of each project:
Query
query {
me {
current_team {
projects {
edges {
node {
title
}
}
}
}
}
}
Response
{
"data": {
"me": {
"current_team": {
"projects": {
"edges": [
{
"node": {
"title": "Test"
}
},
{
"node": {
"title": "Created by the bot"
}
}
]
}
}
}
}
}
Using the team id from that first query, you can create a project:
Query
mutation {
createProject(input: {
clientMutationId: "1",
title: "Created by the bot",
team_id: 526
}) {
project {
dbid
}
}
}
Response
{
"data": {
"createProject": {
"project": {
"dbid": 778
}
}
}
}
When creating medias, it's possible to auto-respond to tasks that are created automatically for any created media. We can configure those tasks. Feel free to suggest those tasks. You can run a query to know which are those tasks, using the project id (from the last query above) and the team id (from the first query above):
Query
query {
project(ids: "778,526") {
auto_tasks
}
}
Response
{
"data": {
"project": {
"auto_tasks": [
{
"label": "When?",
"type": "datetime",
"description": "When did the incident take place",
"projects": [],
"slug": "when"
},
{
"label": "Where?",
"type": "geolocation",
"description": "Where did the incident take place?",
"projects": [],
"slug": "where"
}
]
}
}
}
The slug is what is used to identify the task when creating a media with a response to an auto-task.
The query below adds a new auto-task 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 the choice types, 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: "VGVhbS81MjY=\n",
add_auto_task: "{
\"label\": \"Who?\",
\"type\": \"free_text\",
\"description\": \"\"
}"
}) {
team {
projects(last: 1) {
edges {
node {
auto_tasks
}
}
}
}
}
}
Response
{
"data": {
"updateTeam": {
"team": {
"projects": {
"edges": [
{
"node": {
"auto_tasks": [
{
"label": "When?",
"type": "datetime",
"description": "When did the incident take place",
"projects": [],
"slug": "when"
},
{
"label": "Where?",
"type": "geolocation",
"description": "Where did the incident take place?",
"projects": [],
"slug": "where"
},
{
"label": "Who?",
"type": "free_text",
"description": "",
"slug": "who"
}
]
}
}
]
}
}
}
}
}
You can now add medias (optionally with responses to the auto-tasks) to the project you created before:
Query
mutation {
createProjectMedia(input: {
clientMutationId: "1",
project_id: 878,
quote: "This is a claim",
url: "",
quote_attributions: "{\"name\":\"Source of this claim\"}",
set_tasks_responses: "{
\"when\": \"2017-10-02 15:07 +3 BRT\",
\"where\": \"{\\\"type\\\":\\\"Feature\\\",\\\"geometry\\\":{\\\"type\\\":\\\"Point\\\",\\\"coordinates\\\":[-12.9016241,-38.4198075]},\\\"properties\\\":{\\\"name\\\":\\\"Salvador\\\"}}\"
}"
}) {
project_media {
dbid
}
}
}
Response
{
"data": {
"createProjectMedia": {
"project_media": {
"dbid": 4933
}
}
}
}
As we can see above, the geolocation task response should be a valid GeoJSON.
The url and quote can't be both defined at the same time. It's one or the other. If the quote field is set, it means you're creating a claim. If the url field is set, it means you're creating a report of type "link".
When using createProjectMedia to create new project media, 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
- 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: {
project_id: 4,
set_original_claim: "https://someurl.com/somefile.mp3"
}) {
project_media {
title
type
url
}
}
}This example will create an audio as the original claim.
We can update media using the GraphQL (Base 64) id. In the example below, we change the description of a media:
Query
mutation {
updateProjectMedia(input: {
clientMutationId: "1",
id: "UHJvamVjdE1lZGlhLzY5OTI=\n",
embed: "{\"description\":\"Changing description\"}"
}) {
project_media {
dbid
}
}
}
Response
{
"data": {
"updateProjectMedia": {
"project_media": {
"dbid": 6992
}
}
}
}
You can also add comments to medias:
Query
mutation {
createComment(input: {
clientMutationId: "1",
text: "A comment by a bot",
annotated_id: "4933",
annotated_type: "ProjectMedia"
}) {
comment {
dbid
}
}
}
Response
{
"data": {
"createComment": {
"comment": {
"dbid": "37457"
}
}
}
}
Take a look at GraphiQL interface in order to see all fields that you can read. On the example below, we get the URL, tasks and task responses of all medias marked as verified and that belong to project with database id 778.
Query
query {
search(query: "{\"verification_status\":[\"verified\"],\"projects\":[\"778\"]}") {
medias(first: 10000) {
edges {
node {
tasks(first: 10000) {
edges {
node {
label
first_response_value
}
}
}
media {
url
}
}
}
}
}
}
Response
{
"data": {
"search": {
"medias": {
"edges": [
{
"node": {
"tasks": {
"edges": [
{
"node": {
"label": "Who?",
"first_response_value": "Jaguar"
}
},
{
"node": {
"label": "Where?",
"first_response_value": "Zoo"
}
},
{
"node": {
"label": "When?",
"first_response_value": "A couple weeks ago"
}
}
]
},
"media": {
"url": "https://twitter.com/g1/status/885552426874523648"
}
}
}
]
}
}
}
}
You need to have the id of the media that you want to add a tag to. This is the last number in a Check item URL. For example, for https://checkmedia.org/test/project/73/media/104, the id is 104. That id goes into annotated_id in the query below.
Query
mutation {
createTag(input:{
clientMutationId: "1",
tag: "foo",
annotated_id: "104",
annotated_type: "ProjectMedia"
}) {
tagEdge {
node {
dbid
}
}
}
}
Response
{
"data": {
"createTag": {
"tagEdge": {
"node": {
"dbid": "792871"
}
}
}
}
}
In order to add a related item, you must have the Base 64 id of the child item and the incremental id of the parent item. In the example below, 36064 is the incremental id of the parent item and the Base 64 id of the child item is UHJvamVjdE1lZGlhLzM2MDY3\n.
Query
mutation {
updateProjectMedia(input:{
clientMutationId: "1",
id:"UHJvamVjdE1lZGlhLzM2MDY3\n",
related_to_id: 36064
}) {
related_to {
dbid
}
}
}
Response
{
"data": {
"updateProjectMedia": {
"related_to": {
"dbid": 36064
}
}
}
}