Create a knowledge source to ingest

POST

knowledge

sources

curl --request POST "$MITHUNAI_URL/mithunai/api/v1/knowledge/sources" \
  --header "Authorization: Bearer $MITHUNAI_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "collection_id": "0b6f2c14-8a3d-4e91-9c77-2f5b1d0a4e88",
    "source_type": "website",
    "name": "Product documentation site",
    "configuration": {
      "seed_url": "https://docs.example.com/",
      "path_prefix": "/guides",
      "max_pages": "1000",
      "exclude_paths": "/guides/archive/"
    }
  }'

import os, requests

response = requests.post(
    f"{os.environ['MITHUNAI_URL']}/mithunai/api/v1/knowledge/sources",
    headers={"Authorization": f"Bearer {os.environ['MITHUNAI_API_KEY']}"},
    json={
        "collection_id": "0b6f2c14-8a3d-4e91-9c77-2f5b1d0a4e88",
        "source_type": "website",
        "name": "Product documentation site",
        "configuration": {
            "seed_url": "https://docs.example.com/",
            "path_prefix": "/guides",
            "max_pages": "1000",
            "exclude_paths": "/guides/archive/",
        },
    },
    timeout=60,
)
response.raise_for_status()
print(response.json())

const response = await fetch(`${process.env.MITHUNAI_URL}/mithunai/api/v1/knowledge/sources`, {
  method: 'POST',
  headers: {
    Authorization: `Bearer ${process.env.MITHUNAI_API_KEY}`,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    collection_id: '0b6f2c14-8a3d-4e91-9c77-2f5b1d0a4e88',
    source_type: 'website',
    name: 'Product documentation site',
    configuration: {
      seed_url: 'https://docs.example.com/',
      path_prefix: '/guides',
      max_pages: '1000',
      exclude_paths: '/guides/archive/',
    },
  }),
})
console.log(await response.json())

{
  "id": "5d1e7a90-3c4b-4f2a-8e61-7b9c0d2f4a13",
  "collection_id": "0b6f2c14-8a3d-4e91-9c77-2f5b1d0a4e88",
  "source_type": "website",
  "name": "Product documentation site",
  "configuration": {
    "seed_url": "https://docs.example.com/",
    "path_prefix": "/guides",
    "max_depth": "3",
    "max_pages": "1000",
    "include_paths": "",
    "exclude_paths": "/guides/archive/"
  },
  "status": "active",
  "last_source_version": null,
  "last_ingested_at": null,
  "created_at": "2026-09-24T10:15:02.418331",
  "updated_at": "2026-09-24T10:15:02.418331"
}

{ "code": "validation_error", "message": "The source URL is not valid." }

{ "code": "authorization_error", "message": "You do not have permission to perform this action." }

{ "code": "not_found", "message": "The knowledge collection was not found." }

Creating a source validates and stores its configuration. It does not read anything. To ingest the source, call Sync a source. configuration is a flat object. Every value is a string, and lists are comma-separated inside one string, for example "docs/,README.md". Numbers are accepted and converted to strings. Booleans, arrays and nested objects are refused. The configuration is checked for its source_type when you create the source, so a malformed URL or repository name returns 400 here, not later during ingestion. The response returns the stored form: keys the source type does not read are dropped, and optional keys you left out are filled in with their defaults. No credential can be passed in configuration. Everything in it is readable by every member of your organization. A source that signs in names a saved connection with connection_id instead, and the token stays in the connection. Creating a source requires a role that can manage knowledge. A read-only member receives 403. A collection_id that is not in your organization returns 404. See Knowledge sources for guidance on choosing a source type.

string

required

ID (a UUID) of the collection to ingest into. A malformed ID returns 400. Syncing a source whose collection is archived is accepted, and the job then ends failed.

string

required

One of the source_type values listed on Connectors, such as website, github or upload. Any other value returns 400.

string

required

Display name, 1 to 255 characters after trimming.

object

Settings for the source type. At most 50 keys, and each value at most 8,192 characters, except an upload source’s files, which may hold up to 102,400 (its 200 entries at their longest). The keys each type reads are listed below. The field itself may be omitted, but every source type has a required setting, so in practice a source created without configuration is refused with 400.

Show website

Crawls a documentation site from a seed URL, following links on the same origin (scheme, host and port) only. Redirects are not followed, so a page that redirects is treated as unreadable. URLs ending in image, media, archive, font, stylesheet or script extensions are skipped.

string

required

Absolute http or https URL the crawl starts from, at most 2,048 characters. It must not contain a username or password.

string

default:""

Crawl only paths at or below this prefix, for example /docs. It must start with / and must not contain ... Matching is by path segment, so /docs does not match /docs-internal.

string

default:"3"

Number of links followed from the seed. An integer from 1 to 10.

string

default:"500"

Maximum pages collected in one run. An integer from 1 to 5000.

string

default:""

Comma-separated globs matched against the URL path. When set, only matching pages are ingested. At most 100 globs of up to 512 characters each.

string

default:""

Comma-separated globs matched against the URL path. Matching pages are skipped, even when they also match include_paths. At most 100 globs of up to 512 characters each.

Show sitemap

Reads the URLs listed in a sitemap.xml without following links. It takes the same keys as website. Listed URLs must still be on the sitemap’s origin and pass path_prefix, include_paths and exclude_paths. max_pages caps the pages read. max_depth is validated but not used. A sitemap index is followed to at most 100 child sitemaps.

string

required

Absolute http or https URL of the sitemap, at most 2,048 characters.

Show github

Reads a GitHub repository at a branch, tag or commit. Nothing secret is stored on the source. A public repository is read with the access your deployment is configured with; a private one with the github connection you name in connection_id.

string

required

Account or organization name. Letters, digits, ., _ and -, starting with a letter or digit, not ending in ., at most 100 characters.

string

required

Repository name, with the same rules as owner.

string

default:""

Branch, tag or commit SHA: letters, digits, ., _, / and -, starting with a letter or digit, at most 255 characters. It must not contain .. or //. Leave it empty to follow the repository’s default branch.

string

default:""

Comma-separated globs matched against repository paths. When set, only matching files are ingested. At most 100 globs of up to 512 characters each.

string

Comma-separated globs for files to skip. An exclusion wins over an include. A value you send replaces the default list rather than adding to it. At most 100 globs of up to 512 characters each.

Show upload

Ingests files you stored with Upload files. Post that response’s configuration object back unchanged.

string

required

The manifest: one storage_key|filename entry per file, entries separated by newlines. At most 200 files, with no storage key listed twice. Every storage key must be one issued to your organization.

Show github_issues, github_pull_requests and github_discussions

Reads a repository’s issues, pull requests or discussions, one document each, with comments. All three take the keys below, and are read with the GitHub access your deployment is configured with. github_discussions is in beta and needs that access even for a public repository: without it the source is created and its sync fails.

string

required

Account or organization name, with the same rules as for github.

string

required

Repository name, with the same rules as owner.

string

default:"all"

open, closed or all.

string

default:"365"

Skip anything last updated longer ago than this. An integer from 0 to 3650; 0 means no limit.

string

default:""

Comma-separated labels. When set, only items carrying at least one are read. At most 50 labels of up to 100 characters each.

string

default:""

Comma-separated labels to skip. An exclusion wins over an include. The same bounds as include_labels.

string

default:"200"

How many to read, newest first. An integer from 1 to 1000.

string

default:"30"

Comments read per item. An integer from 0 to 100.

string

default:"true"

github_pull_requests only. "true" skips pull requests closed without being merged.

string

default:"true"

github_pull_requests only. "true" reads the comments on the pull request’s conversation.

string

default:"false"

github_pull_requests only. "true" also reads the review comments left on lines of code.

Show openapi

Reads one public OpenAPI 3.x or Swagger 2.x schema, in JSON or YAML, and produces one document per operation. A $ref to another file or host is not followed, and a schema larger than the deployment’s fetch limit, 1 MiB by default, fails to sync.

string

required

Absolute http or https URL of the schema, at most 2,048 characters. It must not contain a username or password.

string

default:"500"

Maximum operations read. An integer from 1 to 2000.

Show custom_answers

Holds question-and-answer pairs you write, one document per pair. It fetches nothing. A source takes 1 to 25 pairs, and the same question cannot appear twice in it.

string

required

The first question, on one line, at most 500 characters. Further pairs are question_2, question_3 and so on; the numbers set the order.

string

required

The answer to question_1, in Markdown, at most 4,000 characters. Every question needs an answer with the same number.

Show zendesk_help_center

Reads the articles of a Zendesk help centre that a signed-out visitor can read. A help centre that requires signing in fails to sync.

string

required

Any http or https page of the help centre. Only its origin is stored. It must not contain a username or password.

string

default:""

A locale of the help centre, such as en-us. Empty reads its default.

string

default:""

Read one category only. Digits. Cannot be combined with section_id.

string

default:""

Read one section only. Digits. Cannot be combined with category_id.

string

default:"500"

Maximum articles read. An integer from 1 to 2000.

Show discourse

Reads the topics of a public Discourse forum, one document per topic holding its first posts. A forum that requires signing in, or one installed under a path, cannot be read.

string

required

Any http or https page of the forum. Only its origin is stored. It must not contain a username or password.

string

default:""

Read one category only. Digits. Cannot be combined with tag.

string

default:""

Read one tag only: lower-case letters, digits, -, . and _, at most 100 characters. Cannot be combined with category_id.

string

default:"false"

"true" reads only topics with an accepted answer. The sync fails on a forum that does not report accepted answers.

string

default:"100"

Maximum topics read, latest first. An integer from 1 to 500.

string

default:"20"

Posts read from each topic. An integer from 1 to 20.

Show confluence

Reads the current pages of a Confluence Cloud site, one document per page. The site is the one saved with the atlassian connection named in connection_id, which is required; there is no address in the configuration. See Connect Confluence.

string

default:""

Comma-separated space keys, such as ENG,DOCS: letters, digits, ., _ and -, optionally starting with ~, at most 25. Empty reads every space the connection’s account can see.

string

default:"500"

Maximum pages read. An integer from 1 to 5000.

Show notion

Reads the pages and database rows shared with the Notion connection named in connection_id, which is required. See Connect Notion.

string

default:"500"

Maximum pages read. An integer from 1 to 5000.

Show jira

Reads the issues of chosen Jira Cloud projects, one document per issue with its description and comments. Comments restricted to a role or group, and internal service-desk comments, are left out, and no person’s name is kept. The site is the one saved with the atlassian connection named in connection_id, which is required. See Connect Jira.

string

required

Comma-separated project keys, such as ENG,OPS: a letter, then letters, digits or _, at most 25. Lowercase is accepted and stored uppercase.

string

default:"false"

"true" reads only issues whose status is in the Done category.

string

default:"500"

Maximum issues read, oldest first. An integer from 1 to 5000.

Show jira_service_management

Reads the requests of one service desk, one document per request with its public comments only. Internal comments are never read. The site is the one saved with the atlassian connection named in connection_id, which is required. See Connect Jira Service Management.

string

required

The service desk’s number: the digits after /portal/ in its customer portal address.

string

default:"true"

"false" also reads requests that are still open.

string

default:"500"

Maximum requests read. An integer from 1 to 5000.

Show zendesk_tickets

Reads the solved and closed tickets of a Zendesk Support account, one document per ticket with its public replies only. Internal notes are never kept. The account is the one saved with the zendesk connection named in connection_id, which is required. See Connect Zendesk tickets.

string

default:"500"

Maximum tickets read, oldest first. An integer from 1 to 5000.

string

ID (a UUID) of a saved connection to read the source with. Required for confluence, jira and jira_service_management (an atlassian connection), notion (a notion connection) and zendesk_tickets (a zendesk connection); optional for the four GitHub types, where a github connection reads a private repository. Any other type refuses one with 400, as does a connection of the wrong service. A connection in another organization returns 404. You need the owner or admin role to name one. It cannot be changed later, and it is not returned in the response.

boolean

Answer each person only from the documents the source system lets them read. See Source permissions. Offered only where the response’s source_permissions_available is true; true anywhere else returns 400. Left out, a new source takes the default: on wherever it is available.

Sources describes what each type reads and its limits. To change a source after it is created, see Update a source. In include_paths and exclude_paths, * matches within one path segment, ** matches across segments, ? matches one character, and a trailing / matches everything below a directory.

Response

Returns 201 with the new source.

string

required

Source ID (a UUID).

string

required

Collection the source ingests into.

string

required

The source_type the source was created with.

string

required

The trimmed name.

object

required

The stored configuration. Every key the source type reads is present, with defaults filled in, and every value is a string.

string

required

active for a new source.

string | null

required

null until the source has been ingested.

string | null

required

null until the source has been ingested.

string

required

When the source was created, ISO 8601 in UTC.

string

required

When the source was last changed, ISO 8601 in UTC.

curl --request POST "$MITHUNAI_URL/mithunai/api/v1/knowledge/sources" \
  --header "Authorization: Bearer $MITHUNAI_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "collection_id": "0b6f2c14-8a3d-4e91-9c77-2f5b1d0a4e88",
    "source_type": "website",
    "name": "Product documentation site",
    "configuration": {
      "seed_url": "https://docs.example.com/",
      "path_prefix": "/guides",
      "max_pages": "1000",
      "exclude_paths": "/guides/archive/"
    }
  }'

import os, requests

response = requests.post(
    f"{os.environ['MITHUNAI_URL']}/mithunai/api/v1/knowledge/sources",
    headers={"Authorization": f"Bearer {os.environ['MITHUNAI_API_KEY']}"},
    json={
        "collection_id": "0b6f2c14-8a3d-4e91-9c77-2f5b1d0a4e88",
        "source_type": "website",
        "name": "Product documentation site",
        "configuration": {
            "seed_url": "https://docs.example.com/",
            "path_prefix": "/guides",
            "max_pages": "1000",
            "exclude_paths": "/guides/archive/",
        },
    },
    timeout=60,
)
response.raise_for_status()
print(response.json())

const response = await fetch(`${process.env.MITHUNAI_URL}/mithunai/api/v1/knowledge/sources`, {
  method: 'POST',
  headers: {
    Authorization: `Bearer ${process.env.MITHUNAI_API_KEY}`,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    collection_id: '0b6f2c14-8a3d-4e91-9c77-2f5b1d0a4e88',
    source_type: 'website',
    name: 'Product documentation site',
    configuration: {
      seed_url: 'https://docs.example.com/',
      path_prefix: '/guides',
      max_pages: '1000',
      exclude_paths: '/guides/archive/',
    },
  }),
})
console.log(await response.json())

{
  "id": "5d1e7a90-3c4b-4f2a-8e61-7b9c0d2f4a13",
  "collection_id": "0b6f2c14-8a3d-4e91-9c77-2f5b1d0a4e88",
  "source_type": "website",
  "name": "Product documentation site",
  "configuration": {
    "seed_url": "https://docs.example.com/",
    "path_prefix": "/guides",
    "max_depth": "3",
    "max_pages": "1000",
    "include_paths": "",
    "exclude_paths": "/guides/archive/"
  },
  "status": "active",
  "last_source_version": null,
  "last_ingested_at": null,
  "created_at": "2026-09-24T10:15:02.418331",
  "updated_at": "2026-09-24T10:15:02.418331"
}

{ "code": "validation_error", "message": "The source URL is not valid." }

{ "code": "authorization_error", "message": "You do not have permission to perform this action." }

{ "code": "not_found", "message": "The knowledge collection was not found." }

Last modified on November 7, 2026

curl --request POST "$MITHUNAI_URL/mithunai/api/v1/knowledge/sources" \
  --header "Authorization: Bearer $MITHUNAI_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "collection_id": "0b6f2c14-8a3d-4e91-9c77-2f5b1d0a4e88",
    "source_type": "website",
    "name": "Product documentation site",
    "configuration": {
      "seed_url": "https://docs.example.com/",
      "path_prefix": "/guides",
      "max_pages": "1000",
      "exclude_paths": "/guides/archive/"
    }
  }'

import os, requests

response = requests.post(
    f"{os.environ['MITHUNAI_URL']}/mithunai/api/v1/knowledge/sources",
    headers={"Authorization": f"Bearer {os.environ['MITHUNAI_API_KEY']}"},
    json={
        "collection_id": "0b6f2c14-8a3d-4e91-9c77-2f5b1d0a4e88",
        "source_type": "website",
        "name": "Product documentation site",
        "configuration": {
            "seed_url": "https://docs.example.com/",
            "path_prefix": "/guides",
            "max_pages": "1000",
            "exclude_paths": "/guides/archive/",
        },
    },
    timeout=60,
)
response.raise_for_status()
print(response.json())

const response = await fetch(`${process.env.MITHUNAI_URL}/mithunai/api/v1/knowledge/sources`, {
  method: 'POST',
  headers: {
    Authorization: `Bearer ${process.env.MITHUNAI_API_KEY}`,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    collection_id: '0b6f2c14-8a3d-4e91-9c77-2f5b1d0a4e88',
    source_type: 'website',
    name: 'Product documentation site',
    configuration: {
      seed_url: 'https://docs.example.com/',
      path_prefix: '/guides',
      max_pages: '1000',
      exclude_paths: '/guides/archive/',
    },
  }),
})
console.log(await response.json())

{
  "id": "5d1e7a90-3c4b-4f2a-8e61-7b9c0d2f4a13",
  "collection_id": "0b6f2c14-8a3d-4e91-9c77-2f5b1d0a4e88",
  "source_type": "website",
  "name": "Product documentation site",
  "configuration": {
    "seed_url": "https://docs.example.com/",
    "path_prefix": "/guides",
    "max_depth": "3",
    "max_pages": "1000",
    "include_paths": "",
    "exclude_paths": "/guides/archive/"
  },
  "status": "active",
  "last_source_version": null,
  "last_ingested_at": null,
  "created_at": "2026-09-24T10:15:02.418331",
  "updated_at": "2026-09-24T10:15:02.418331"
}

{ "code": "validation_error", "message": "The source URL is not valid." }

{ "code": "authorization_error", "message": "You do not have permission to perform this action." }

{ "code": "not_found", "message": "The knowledge collection was not found." }

Get started

Conversations

Assistants

Models

Knowledge

Widget deployments

Widget (public)

MCP

Connections

Slack

Microsoft Teams

API keys

Members

Invitations

Analytics

Support access

Audit log

AI search visibility

AI employees

Branding

Operations

Create a knowledge source to ingest

Response

​Response

Response