Create Learnings
curl --request POST \
--url https://api.coderabbit.ai/v1/knowledge/learnings \
--header 'Content-Type: application/json' \
--header 'x-coderabbitai-api-key: <api-key>' \
--data '
{
"learnings": [
{
"repoName": "<string>",
"repoOwner": "<string>",
"learning": "<string>",
"user": "<string>",
"file": "<string>",
"url": "<string>",
"userUID": "<string>",
"pullRequest": 123,
"startLine": 123,
"endLine": 123,
"topLevelCommentId": "<string>"
}
],
"org_id": "<string>",
"org_name": "<string>"
}
'import requests
url = "https://api.coderabbit.ai/v1/knowledge/learnings"
payload = {
"learnings": [
{
"repoName": "<string>",
"repoOwner": "<string>",
"learning": "<string>",
"user": "<string>",
"file": "<string>",
"url": "<string>",
"userUID": "<string>",
"pullRequest": 123,
"startLine": 123,
"endLine": 123,
"topLevelCommentId": "<string>"
}
],
"org_id": "<string>",
"org_name": "<string>"
}
headers = {
"x-coderabbitai-api-key": "<api-key>",
"Content-Type": "application/json"
}
response = requests.post(url, json=payload, headers=headers)
print(response.text)const options = {
method: 'POST',
headers: {'x-coderabbitai-api-key': '<api-key>', 'Content-Type': 'application/json'},
body: JSON.stringify({
learnings: [
{
repoName: '<string>',
repoOwner: '<string>',
learning: '<string>',
user: '<string>',
file: '<string>',
url: '<string>',
userUID: '<string>',
pullRequest: 123,
startLine: 123,
endLine: 123,
topLevelCommentId: '<string>'
}
],
org_id: '<string>',
org_name: '<string>'
})
};
fetch('https://api.coderabbit.ai/v1/knowledge/learnings', options)
.then(res => res.json())
.then(res => console.log(res))
.catch(err => console.error(err));{
"success": true,
"learning_ids": [
"3c90c3cc-0d44-4b50-8888-8dd25736052a"
]
}{
"error": {
"code": "invalid_request",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "unauthenticated",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "forbidden",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "not_found",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "head_changed",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "precondition_failed",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "payload_too_large",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "rate_limit_exceeded",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "internal_error",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "unavailable",
"message": "<string>",
"retry_after": 1
}
}Create Learnings
Creates one Learning for each submitted item. The request has no Learning ID, so repeating it can create duplicates. Returns the created IDs. Read each with GET /knowledge/learnings/. learning_ids is null when an older serving writer cannot return identity metadata; do not replay the write to discover IDs.
POST
/
v1
/
knowledge
/
learnings
Create Learnings
curl --request POST \
--url https://api.coderabbit.ai/v1/knowledge/learnings \
--header 'Content-Type: application/json' \
--header 'x-coderabbitai-api-key: <api-key>' \
--data '
{
"learnings": [
{
"repoName": "<string>",
"repoOwner": "<string>",
"learning": "<string>",
"user": "<string>",
"file": "<string>",
"url": "<string>",
"userUID": "<string>",
"pullRequest": 123,
"startLine": 123,
"endLine": 123,
"topLevelCommentId": "<string>"
}
],
"org_id": "<string>",
"org_name": "<string>"
}
'import requests
url = "https://api.coderabbit.ai/v1/knowledge/learnings"
payload = {
"learnings": [
{
"repoName": "<string>",
"repoOwner": "<string>",
"learning": "<string>",
"user": "<string>",
"file": "<string>",
"url": "<string>",
"userUID": "<string>",
"pullRequest": 123,
"startLine": 123,
"endLine": 123,
"topLevelCommentId": "<string>"
}
],
"org_id": "<string>",
"org_name": "<string>"
}
headers = {
"x-coderabbitai-api-key": "<api-key>",
"Content-Type": "application/json"
}
response = requests.post(url, json=payload, headers=headers)
print(response.text)const options = {
method: 'POST',
headers: {'x-coderabbitai-api-key': '<api-key>', 'Content-Type': 'application/json'},
body: JSON.stringify({
learnings: [
{
repoName: '<string>',
repoOwner: '<string>',
learning: '<string>',
user: '<string>',
file: '<string>',
url: '<string>',
userUID: '<string>',
pullRequest: 123,
startLine: 123,
endLine: 123,
topLevelCommentId: '<string>'
}
],
org_id: '<string>',
org_name: '<string>'
})
};
fetch('https://api.coderabbit.ai/v1/knowledge/learnings', options)
.then(res => res.json())
.then(res => console.log(res))
.catch(err => console.error(err));{
"success": true,
"learning_ids": [
"3c90c3cc-0d44-4b50-8888-8dd25736052a"
]
}{
"error": {
"code": "invalid_request",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "unauthenticated",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "forbidden",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "not_found",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "head_changed",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "precondition_failed",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "payload_too_large",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "rate_limit_exceeded",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "internal_error",
"message": "<string>",
"retry_after": 1
}
}{
"error": {
"code": "unavailable",
"message": "<string>",
"retry_after": 1
}
}- CLI Tool
- MCP Server
coderabbit api request /knowledge/learnings --input request.json
An agent connected to the CodeRabbit MCP server runs this operation as the
create_learnings tool. The connection selects the organization, so the arguments have no org_id or org_name.{
"method": "tools/call",
"params": {
"name": "create_learnings",
"arguments": {
"learnings": []
}
}
}
Authorizations
See Authentication for the key types and where to create a key.
Body
application/json
Organization for a workspace key: the Git-provider organization ID (scope.provider_organization_id) or the CodeRabbit organization ID (scope.id, or id from list organizations). Supply either org_id or org_name, never both. Omit for organization keys and MCP connections.
Minimum string length:
1Was this page helpful?