Full Text Search

Perform full text search for exact matches within your indexed sources.

Use this endpoint to find content chunks using BM25-based text matching with configurable operators. Choose between ‘OR’ and ‘AND’ operators to control how search terms are combined for precise text matching.

Hit the Try it button to try this API now in our playground. It’s the best way to check the full request and response in one place, customize your parameters, and generate ready-to-use code snippets.

Examples

API Request

curl --request POST \
  --url https://api.usecortex.ai/search/full-text-search \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "query": "John Smith Jake",
  "tenant_id": "tenant_1234",
  "sub_tenant_id": "sub_tenant_4567",
  "max_chunks": 25,
  "operator": "and"
}'

Search Options

The Full Text Search endpoint provides various operators and search capabilities to help you find exactly what you’re looking for in your indexed content.

Boolean Operators

AND Operator

• Usage: "operator": "and"
• Behavior: All search terms must be present in the document for it to match
• Best for: Precise searches where you need all keywords to be present
• Example: Searching for “machine learning algorithms” with AND will only return documents containing both “machine” AND “learning” AND “algorithms”

OR Operator

• Usage: "operator": "or"
• Behavior: At least one search term must be present in the document for it to match
• Best for: Broad searches to find documents containing any of the specified terms
• Example: Searching for “python javascript react” with OR will return documents containing any of these programming languages

Search Optimization Tips

For Better Precision

• Use the AND operator when you need all terms to be present
• Use specific terminology rather than generic terms

For Broader Results

• Use the OR operator to find documents with any of the search terms
• Try synonyms and related terms to expand your search

Response

Returns an array of relevant content chunks from your indexed sources based on the full text search query.

[
  {
    "chunk_uuid": "CortexDoc37e854b429784b148d4fc910812bdc581753761779_20_v1",
    "source_id": "CortexDoc37e854b429784b148d4fc910812bdc581753761779",
    "source_title": "IEEE Transactions LaTeX Templates.pdf",
    "chunk_content": "JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 12 Fig. 10: Comparison of models with 4bit precision on Medical QA dataset...",
    "source_url": "ToayuCogoBdxZVoZQ1ft8ZoRzCO2/Hello/Hello/local_source/CortexDoc37e854b429784b148d4fc910812bdc581753761779",
    "source_upload_time": "1753761802.079415",
    "source_collection": [],
    "source_type": "file",
    "layout": "{\"coordinates\": {\"x\": 48.96399688720703, \"y\": 26.49277114868164, \"width\": 514.0717697143555, \"height\": 721.1119651794434}, \"page\": 12}",
    "version": "v1",
    "source_last_updated_time": "",
    "relevancy_score": 0.8363813161849976,
    "rerank_score": null
  },
  {
    "chunk_uuid": "CortexDoc04d73ba38fd54325800a74ec17aed9041753761993_4_v1",
    "source_id": "CortexDoc04d73ba38fd54325800a74ec17aed9041753761993",
    "source_title": "IEEE Transactions LaTeX Templates.pdf",
    "chunk_content": "III. METHODOLOGY Our proposed framework integrates federated learning (FL) with blockchain technology to fine-tune Large Language Mod-",
    "source_url": "ToayuCogoBdxZVoZQ1ft8ZoRzCO2/Hello/Hello/local_source/CortexDoc04d73ba38fd54325800a74ec17aed9041753761993",
    "source_upload_time": "1753762013.1677928",
    "source_collection": [],
    "source_type": "file",
    "layout": "{\"coordinates\": {\"x\": 311.9779968261719, \"y\": 712.3123168945312, \"width\": 251.05746459960938, \"height\": 35.72357177734375}, \"page\": 2}",
    "version": "v1",
    "source_last_updated_time": "",
    "relevancy_score": 0.800000011920929,
    "rerank_score": null
  }
]

Note: This API performs full text search with configurable operators for precise text matching. For conversational Q&A with AI-generated responses, use the /search/qna API instead.

Operator Parameter

The operator parameter controls how the search terms are combined:

• OR operator (default): At least one token must be present in the document
• AND operator: All tokens must be present in the document

Max Chunks Parameter

The max_chunks parameter controls the maximum number of results returned:

• Must be between 1 and 1001
• Defaults to the system limit if not specified

Use Cases

• Precise keyword matching: Use AND operator when you need all search terms to be present
• Broad search: Use OR operator to find documents containing any of the search terms
• Exact phrase matching: Ideal for finding specific terminology or phrases in your documents

Error Responses

All endpoints return consistent error responses following the standard format. For detailed error information, see our Error Responses documentation.

Authorizations

Authorization · string · header · required

Body

application/json

query · string · required

Example:

tenant_id · string · required

Example:

sub_tenant_id · string | null · default: ""

Example:

operator · enum<string>

Available options:

or,

and

max_chunks · integer | null · default: 25

Response

200

application/json

chunk_uuid · string · required

Example:

source_id · string · required

Example:

chunk_content · string · required

Example:

source_type · string · default: ""

Example:

source_upload_time · string · default: ""

Example:

source_title · string · default: ""

Example:

source_last_updated_time · string · default: ""

Example:

layout · string | null

Example:

relevancy_score · number | null

document_metadata · object | null

Example:

tenant_metadata · object | null

Example:

extra_context · object | null

graph_triplet_ids · string[]

Example: