Get Config

Select a knowledge base

Register a Knowledge with contents_db on AgentOS, an Agent or a Team. Read the actual generated ID from GET /config under knowledge.knowledge_instances; use that knowledge_id in the request's selector. It is not an arbitrary Knowledge.id. A selector is required when multiple instances make the request ambiguous (400); no configured knowledge base returns 503. If both selectors are supplied, knowledge_id takes precedence over db_id.

Ingestion and search also require suitable readers, vector storage, an embedder where needed, and their dependencies and credentials. See Knowledge.

Choose reader and chunker IDs from the current response. Availability depends on installed packages; unavailable_readers and unavailable_chunkers identify missing requirements. The long generated example is illustrative and can be abbreviated by the renderer. Configured remote sources describe ingestion sources; they are distinct from a RemoteKnowledge instance.

GET/knowledge/config

Retrieve available readers, chunkers, and configuration options for content processing. This endpoint provides metadata about supported file types, processing strategies, and filters.

Authorization

HTTPBearer
AuthorizationBearer <token>

In: header

Query Parameters

db_id?|

Database ID to use

knowledge_id?|

Knowledge base ID to use

Response Body

application/json

application/json

application/json

application/json

application/json

application/json

curl --request GET 'https://example.com/knowledge/config'
{  "readers": {    "website": {      "id": "website",      "name": "WebsiteReader",      "description": "Reads website files",      "chunkers": [        "AgenticChunker",        "DocumentChunker",        "RecursiveChunker",        "SemanticChunker",        "FixedSizeChunker"      ]    },    "firecrawl": {      "id": "firecrawl",      "name": "FirecrawlReader",      "description": "Reads firecrawl files",      "chunkers": [        "SemanticChunker",        "FixedSizeChunker",        "AgenticChunker",        "DocumentChunker",        "RecursiveChunker"      ]    },    "youtube": {      "id": "youtube",      "name": "YoutubeReader",      "description": "Reads youtube files",      "chunkers": [        "RecursiveChunker",        "AgenticChunker",        "DocumentChunker",        "SemanticChunker",        "FixedSizeChunker"      ]    },    "web_search": {      "id": "web_search",      "name": "WebSearchReader",      "description": "Reads web_search files",      "chunkers": [        "AgenticChunker",        "DocumentChunker",        "RecursiveChunker",        "SemanticChunker",        "FixedSizeChunker"      ]    },    "arxiv": {      "id": "arxiv",      "name": "ArxivReader",      "description": "Reads arxiv files",      "chunkers": [        "FixedSizeChunker",        "AgenticChunker",        "DocumentChunker",        "RecursiveChunker",        "SemanticChunker"      ]    },    "csv": {      "id": "csv",      "name": "CsvReader",      "description": "Reads csv files",      "chunkers": [        "RowChunker",        "FixedSizeChunker",        "AgenticChunker",        "DocumentChunker",        "RecursiveChunker"      ]    },    "docling": {      "id": "docling",      "name": "DoclingReader",      "description": "Converts multiple document formats like PDF, DOCX, PPTX, images, HTML, etc. using IBM's Docling library",      "chunkers": [        "AgenticChunker",        "CodeChunker",        "DocumentChunker",        "FixedSizeChunker",        "RecursiveChunker",        "SemanticChunker"      ]    },    "docx": {      "id": "docx",      "name": "DocxReader",      "description": "Reads docx files",      "chunkers": [        "DocumentChunker",        "FixedSizeChunker",        "SemanticChunker",        "AgenticChunker",        "RecursiveChunker"      ]    },    "gcs": {      "id": "gcs",      "name": "GcsReader",      "description": "Reads gcs files",      "chunkers": [        "FixedSizeChunker",        "AgenticChunker",        "DocumentChunker",        "RecursiveChunker",        "SemanticChunker"      ]    },    "json": {      "id": "json",      "name": "JsonReader",      "description": "Reads json files",      "chunkers": [        "FixedSizeChunker",        "AgenticChunker",        "DocumentChunker",        "RecursiveChunker",        "SemanticChunker"      ]    },    "markdown": {      "id": "markdown",      "name": "MarkdownReader",      "description": "Reads markdown files",      "chunkers": [        "MarkdownChunker",        "DocumentChunker",        "AgenticChunker",        "RecursiveChunker",        "SemanticChunker",        "FixedSizeChunker"      ]    },    "pdf": {      "id": "pdf",      "name": "PdfReader",      "description": "Reads pdf files",      "chunkers": [        "DocumentChunker",        "FixedSizeChunker",        "AgenticChunker",        "SemanticChunker",        "RecursiveChunker"      ]    },    "text": {      "id": "text",      "name": "TextReader",      "description": "Reads text files",      "chunkers": [        "CodeChunker",        "FixedSizeChunker",        "AgenticChunker",        "DocumentChunker",        "RecursiveChunker",        "SemanticChunker"      ]    }  },  "readersForType": {    "url": [      "url",      "website",      "firecrawl",      "youtube",      "web_search",      "gcs"    ],    "youtube": [      "youtube"    ],    "text": [      "web_search"    ],    "topic": [      "arxiv"    ],    "file": [      "csv",      "gcs"    ],    ".csv": [      "csv",      "field_labeled_csv",      "docling"    ],    ".xlsx": [      "excel",      "docling"    ],    ".xls": [      "excel"    ],    "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet": [      "excel"    ],    "application/vnd.ms-excel": [      "excel"    ],    ".docx": [      "docx",      "docling"    ],    ".dotx": [      "docling"    ],    ".docm": [      "docling"    ],    ".dotm": [      "docling"    ],    ".pptx": [      "docling",      "pptx"    ],    ".potx": [      "docling"    ],    ".ppsx": [      "docling"    ],    ".pptm": [      "docling"    ],    ".ppsm": [      "docling"    ],    ".potm": [      "docling"    ],    ".json": [      "json"    ],    ".md": [      "markdown",      "docling"    ],    ".pdf": [      "pdf",      "docling"    ],    ".txt": [      "text"    ],    ".html": [      "docling"    ],    ".htm": [      "docling"    ],    ".xhtml": [      "docling"    ],    ".xml": [      "docling"    ],    ".nxml": [      "docling"    ],    ".xbrl": [      "docling"    ],    ".adoc": [      "docling"    ],    ".asciidoc": [      "docling"    ],    ".asc": [      "docling"    ],    ".xlsm": [      "docling"    ],    ".tex": [      "docling"    ],    ".latex": [      "docling"    ],    ".tar.gz": [      "docling"    ],    ".vtt": [      "docling"    ],    ".png": [      "docling"    ],    ".jpeg": [      "docling"    ],    ".jpg": [      "docling"    ],    ".tiff": [      "docling"    ],    ".tif": [      "docling"    ],    ".bmp": [      "docling"    ],    ".webp": [      "docling"    ],    ".wav": [      "docling"    ],    ".mp3": [      "docling"    ],    ".m4a": [      "docling"    ],    ".aac": [      "docling"    ],    ".ogg": [      "docling"    ],    ".flac": [      "docling"    ],    ".mp4": [      "docling"    ],    ".avi": [      "docling"    ],    ".mov": [      "docling"    ]  },  "chunkers": {    "AgenticChunker": {      "key": "AgenticChunker",      "name": "AgenticChunker",      "description": "Chunking strategy that uses an LLM to determine natural breakpoints in the text",      "metadata": {        "chunk_size": 5000      }    },    "CodeChunker": {      "key": "CodeChunker",      "name": "CodeChunker",      "description": "The CodeChunker splits code into chunks based on its structure, leveraging Abstract Syntax Trees (ASTs) to create contextually relevant segments",      "metadata": {        "chunk_size": 2048      }    },    "DocumentChunker": {      "key": "DocumentChunker",      "name": "DocumentChunker",      "description": "A chunking strategy that splits text based on document structure like paragraphs and sections",      "metadata": {        "chunk_size": 5000,        "chunk_overlap": 0      }    },    "FixedSizeChunker": {      "key": "FixedSizeChunker",      "name": "FixedSizeChunker",      "description": "Chunking strategy that splits text into fixed-size chunks with optional overlap",      "metadata": {        "chunk_size": 5000,        "chunk_overlap": 0      }    },    "MarkdownChunker": {      "key": "MarkdownChunker",      "name": "MarkdownChunker",      "description": "A chunking strategy that splits markdown based on structure like headers, paragraphs and sections",      "metadata": {        "chunk_size": 5000,        "chunk_overlap": 0      }    },    "RecursiveChunker": {      "key": "RecursiveChunker",      "name": "RecursiveChunker",      "description": "Chunking strategy that recursively splits text into chunks by finding natural break points",      "metadata": {        "chunk_size": 5000,        "chunk_overlap": 0      }    },    "RowChunker": {      "key": "RowChunker",      "name": "RowChunker",      "description": "RowChunking chunking strategy",      "metadata": {}    },    "SemanticChunker": {      "key": "SemanticChunker",      "name": "SemanticChunker",      "description": "Chunking strategy that splits text into semantic chunks using chonkie",      "metadata": {        "chunk_size": 5000      }    }  },  "vector_dbs": [    {      "id": "vector_db_1",      "name": "Vector DB 1",      "description": "Vector DB 1 description",      "search_types": [        "vector",        "keyword",        "hybrid"      ]    }  ],  "filters": [    "filter_tag_1",    "filter_tag2"  ]}