Skip to content

Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki.

24.1 KB
Raw Blame
// Package client provides the implementation for interacting with the WeKnora API
// The KnowledgeBase related interfaces are used to manage knowledge bases
// Knowledge bases are collections of knowledge entries that can be used for question-answering
// They can also be searched and queried using hybrid search
package client

import (
	"context"
	"encoding/json"
	"fmt"
	"net/http"
	"net/url"
	"time"
)

// KnowledgeBase represents a knowledge base
type KnowledgeBase struct {
	ID                    string                 `json:"id"`
	Name                  string                 `json:"name"` // Name must be unique within the same tenant
	Type                  string                 `json:"type"`
	IsTemporary           bool                   `json:"is_temporary"`
	IsPinned              bool                   `json:"is_pinned"`
	Description           string                 `json:"description"`
	TenantID              uint64                 `json:"tenant_id"`
	ChunkingConfig        ChunkingConfig         `json:"chunking_config"`
	ImageProcessingConfig ImageProcessingConfig  `json:"image_processing_config"`
	FAQConfig             *FAQConfig             `json:"faq_config"`
	EmbeddingModelID      string                 `json:"embedding_model_id"`
	SummaryModelID        string                 `json:"summary_model_id"`
	VLMConfig             VLMConfig              `json:"vlm_config"`
	StorageProviderConfig *StorageProviderConfig `json:"storage_provider_config"`
	StorageConfig         StorageConfig          `json:"storage_config"`
	ExtractConfig         *ExtractConfig         `json:"extract_config"`
	AutoTagConfig         *AutoTagConfig         `json:"auto_tag_config"`
	CreatedAt             time.Time              `json:"created_at"`
	UpdatedAt             time.Time              `json:"updated_at"`
	// Computed fields (not stored in database)
	KnowledgeCount  int64 `json:"knowledge_count"`
	ChunkCount      int64 `json:"chunk_count"`
	IsProcessing    bool  `json:"is_processing"`
	ProcessingCount int64 `json:"processing_count"`
}

// KnowledgeBaseConfig represents knowledge base configuration
type KnowledgeBaseConfig struct {
	ChunkingConfig        ChunkingConfig        `json:"chunking_config"`
	ImageProcessingConfig ImageProcessingConfig `json:"image_processing_config"`
	FAQConfig             *FAQConfig            `json:"faq_config"`
	AutoTagConfig         *AutoTagConfig        `json:"auto_tag_config,omitempty"`
}

// ChunkingConfig represents document chunking configuration
type ChunkingConfig struct {
	ChunkSize    int      `json:"chunk_size"`    // Chunk size
	ChunkOverlap int      `json:"chunk_overlap"` // Overlap size
	Separators   []string `json:"separators"`    // Separators
	// ParserEngineRules selects the parser engine for each group of file types.
	ParserEngineRules []ParserEngineRule `json:"parser_engine_rules,omitempty"`
}

// IndexingStrategy selects which indexing pipelines are active.
type IndexingStrategy struct {
	VectorEnabled  bool `json:"vector_enabled"`
	KeywordEnabled bool `json:"keyword_enabled"`
	WikiEnabled    bool `json:"wiki_enabled"`
	GraphEnabled   bool `json:"graph_enabled"`
}

// FAQConfig represents faq-specific configuration
type FAQConfig struct {
	IndexMode         string `json:"index_mode"`
	QuestionIndexMode string `json:"question_index_mode"`
}

// ImageProcessingConfig represents image processing configuration
type ImageProcessingConfig struct {
	ModelID string `json:"model_id"` // Multimodal model ID
}

// VLMConfig represents the VLM configuration
type VLMConfig struct {
	Enabled bool   `json:"enabled"`
	ModelID string `json:"model_id"`
}

// StorageProviderConfig stores the KB-level storage provider selection.
type StorageProviderConfig struct {
	Provider string `json:"provider"`
}

// StorageConfig represents the legacy storage configuration (cos_config).
// Deprecated: use StorageProviderConfig for provider selection.
type StorageConfig struct {
	SecretID   string `json:"secret_id"`
	SecretKey  string `json:"secret_key"`
	Region     string `json:"region"`
	BucketName string `json:"bucket_name"`
	AppID      string `json:"app_id"`
	PathPrefix string `json:"path_prefix"`
	Provider   string `json:"provider"`
}

// ExtractConfig represents the extract configuration for a knowledge base
type ExtractConfig struct {
	Enabled   bool             `json:"enabled"`
	Text      string           `json:"text,omitempty"`
	Tags      []string         `json:"tags,omitempty"`
	Nodes     []*GraphNode     `json:"nodes,omitempty"`
	Relations []*GraphRelation `json:"relations,omitempty"`
}

// GraphNode represents a node in the graph extraction configuration
type GraphNode struct {
	Name string `json:"name"`
}

// GraphRelation represents a relation in the graph extraction configuration
type GraphRelation struct {
	Node1 string `json:"node1"`
	Node2 string `json:"node2"`
	Type  string `json:"type"`
}

// ParserEngineRule maps a set of file types to a specific parser engine.
type ParserEngineRule struct {
	FileTypes []string `json:"file_types"`
	Engine    string   `json:"engine"`
}

// QuestionGenerationConfig controls LLM-generated questions per chunk during parsing.
type QuestionGenerationConfig struct {
	Enabled       bool `json:"enabled"`
	QuestionCount int  `json:"question_count"`
}

// AutoTagConfig controls optional automatic association of existing knowledge
// base tags after a document finishes parsing. Only applies to document-type
// knowledge bases; disabled by default.
type AutoTagConfig struct {
	Enabled bool `json:"enabled"`
	// ModelID selects the chat model used for classification. Empty falls
	// back to the knowledge base's summary model.
	ModelID string `json:"model_id,omitempty"`
	// MaxTags caps how many existing tags one document may auto-acquire
	// (1-10, defaults to 3).
	MaxTags int `json:"max_tags,omitempty"`
	// SkipIfTagged leaves documents that already carry tags untouched.
	// Defaults to true when omitted.
	SkipIfTagged *bool `json:"skip_if_tagged,omitempty"`
}

// ASRConfig represents automatic speech recognition settings for audio files.
type ASRConfig struct {
	Enabled  bool   `json:"enabled"`
	ModelID  string `json:"model_id"`
	Language string `json:"language,omitempty"`
}

// UnmarshalJSON keeps backward compatibility for legacy responses that still
// use `cos_config` instead of `storage_config`.
func (kb *KnowledgeBase) UnmarshalJSON(data []byte) error {
	type alias KnowledgeBase
	aux := struct {
		*alias
		LegacyStorageConfig *StorageConfig `json:"cos_config"`
	}{
		alias: (*alias)(kb),
	}
	if err := json.Unmarshal(data, &aux); err != nil {
		return err
	}
	if aux.LegacyStorageConfig != nil && kb.StorageConfig == (StorageConfig{}) {
		kb.StorageConfig = *aux.LegacyStorageConfig
	}
	return nil
}

// KnowledgeBaseResponse knowledge base response
type KnowledgeBaseResponse struct {
	Success bool          `json:"success"`
	Data    KnowledgeBase `json:"data"`
}

// KnowledgeBaseListResponse knowledge base list response
type KnowledgeBaseListResponse struct {
	Success bool            `json:"success"`
	Data    []KnowledgeBase `json:"data"`
}

// MatchType records which retrieval channel produced a SearchResult.
// Numeric values are the wire contract; they mirror the iota order in
// server-side internal/types/embedding.go — do not reorder without
// coordinating a server bump. Server-side names are preserved as
// trailing comments so cross-repo grep works in both directions.
//
// Channel grouping: 0-1 primary text channels (vector + keyword);
// 2-5 enrichment chunks (added in addition to primary matches, score=0);
// 6-9 alternate sources (graph DB, web search, raw load, data analysis).
type MatchType int

const (
	MatchTypeVector   MatchType = 0 // server: MatchTypeEmbedding
	MatchTypeKeyword  MatchType = 1 // server: MatchTypeKeywords
	MatchTypeNearby   MatchType = 2 // server: MatchTypeNearByChunk
	MatchTypeHistory  MatchType = 3 // server: MatchTypeHistory
	MatchTypeParent   MatchType = 4 // server: MatchTypeParentChunk
	MatchTypeRelation MatchType = 5 // server: MatchTypeRelationChunk
	MatchTypeGraph    MatchType = 6 // server: MatchTypeGraph
	MatchTypeWeb      MatchType = 7 // server: MatchTypeWebSearch
	MatchTypeDirect   MatchType = 8 // server: MatchTypeDirectLoad — chunk loaded by ID without scoring
	MatchTypeData     MatchType = 9 // server: MatchTypeDataAnalysis — produced by analytical pipeline, not retrieval
)

// SearchResult represents search result.
//
// Score is the RRF (reciprocal-rank-fusion) score combining vector and
// keyword channels — typically in the [0, ~0.03] range when both channels
// hit, NOT the raw vector similarity. Use MatchType to tell which channel
// produced each result. Per-channel thresholds (vector_threshold,
// keyword_threshold) filter pre-fusion at retrieval time, before RRF runs.
type SearchResult struct {
	ID                string            `json:"id"`
	Content           string            `json:"content"`
	KnowledgeID       string            `json:"knowledge_id"`
	ChunkIndex        int               `json:"chunk_index"`
	KnowledgeTitle    string            `json:"knowledge_title"`
	StartAt           int               `json:"start_at"`
	EndAt             int               `json:"end_at"`
	Seq               int               `json:"seq"`
	Score             float64           `json:"score"`
	MatchType         MatchType         `json:"match_type"`
	ChunkType         string            `json:"chunk_type"`
	ImageInfo         string            `json:"image_info"`
	Metadata          map[string]string `json:"metadata"`
	KnowledgeFilename string            `json:"knowledge_filename"`
	KnowledgeSource   string            `json:"knowledge_source"`
	KnowledgeChannel  string            `json:"knowledge_channel"`
	// MatchedContent is the actual content that was matched in vector search
	// For FAQ: this is the matched question text (standard or similar question)
	MatchedContent string `json:"matched_content,omitempty"`
	// KnowledgeBaseID identifies the KB that produced this hit. It is required
	// for agent runs that search more than one knowledge base.
	KnowledgeBaseID string `json:"knowledge_base_id,omitempty"`
	// ParentChunkID / SubChunkID describe the chunk hierarchy: a retrieval
	// "hit" id may be a sub-chunk, while the parent holds the self-contained
	// passage an agent fetches via `chunk view <parent_chunk_id>`. The server
	// ships these in the references event; without fields here Go silently
	// drops them during unmarshal.
	ParentChunkID string   `json:"parent_chunk_id,omitempty"`
	SubChunkID    []string `json:"sub_chunk_id,omitempty"`
}

// HybridSearchResponse hybrid search response
type HybridSearchResponse struct {
	Success bool            `json:"success"`
	Data    []*SearchResult `json:"data"`
	// Meta is present when the request carried a rerank object.
	Meta *RetrievalMeta `json:"meta,omitempty"`
}

type CopyKnowledgeBaseRequest struct {
	TaskID   string `json:"task_id,omitempty"`
	SourceID string `json:"source_id"`
	TargetID string `json:"target_id"`
}

// CopyKnowledgeBaseResponse represents the response from copy knowledge base API
type CopyKnowledgeBaseResponse struct {
	TaskID   string `json:"task_id"`
	SourceID string `json:"source_id"`
	TargetID string `json:"target_id"`
	Message  string `json:"message"`
}

type DuplicateKnowledgeBaseResponse struct {
	SourceID      string        `json:"source_id"`
	TargetID      string        `json:"target_id"`
	Message       string        `json:"message"`
	KnowledgeBase KnowledgeBase `json:"knowledge_base"`
}

// KBCloneProgress represents the progress of a knowledge base clone task
type KBCloneProgress struct {
	TaskID    string `json:"task_id"`
	SourceID  string `json:"source_id"`
	TargetID  string `json:"target_id"`
	Status    string `json:"status"`    // pending, processing, completed, failed
	Progress  int    `json:"progress"`  // 0-100
	Total     int    `json:"total"`     // Total operations count
	Processed int    `json:"processed"` // Processed operations count
	Message   string `json:"message"`
	Error     string `json:"error,omitempty"`
	CreatedAt int64  `json:"created_at"`
	UpdatedAt int64  `json:"updated_at"`
}

// CreateKnowledgeBase creates a knowledge base
func (c *Client) CreateKnowledgeBase(ctx context.Context, knowledgeBase *KnowledgeBase) (*KnowledgeBase, error) {
	resp, err := c.doRequest(ctx, http.MethodPost, "/api/v1/knowledge-bases", knowledgeBase, nil)
	if err != nil {
		return nil, err
	}

	var response KnowledgeBaseResponse
	if err := parseResponse(resp, &response); err != nil {
		return nil, err
	}

	return &response.Data, nil
}

// GetKnowledgeBase gets a knowledge base
func (c *Client) GetKnowledgeBase(ctx context.Context, knowledgeBaseID string) (*KnowledgeBase, error) {
	path := fmt.Sprintf("/api/v1/knowledge-bases/%s", knowledgeBaseID)
	resp, err := c.doRequest(ctx, http.MethodGet, path, nil, nil)
	if err != nil {
		return nil, err
	}

	var response KnowledgeBaseResponse
	if err := parseResponse(resp, &response); err != nil {
		return nil, err
	}

	return &response.Data, nil
}

// ListKnowledgeBases lists knowledge bases
func (c *Client) ListKnowledgeBases(ctx context.Context) ([]KnowledgeBase, error) {
	resp, err := c.doRequest(ctx, http.MethodGet, "/api/v1/knowledge-bases", nil, nil)
	if err != nil {
		return nil, err
	}

	var response KnowledgeBaseListResponse
	if err := parseResponse(resp, &response); err != nil {
		return nil, err
	}

	return response.Data, nil
}

// UpdateKnowledgeBaseRequest update knowledge base request
type UpdateKnowledgeBaseRequest struct {
	Name        string               `json:"name"`
	Description string               `json:"description"`
	Config      *KnowledgeBaseConfig `json:"config"`
}

// UpdateKnowledgeBase updates a knowledge base
func (c *Client) UpdateKnowledgeBase(ctx context.Context,
	knowledgeBaseID string,
	request *UpdateKnowledgeBaseRequest,
) (*KnowledgeBase, error) {
	path := fmt.Sprintf("/api/v1/knowledge-bases/%s", knowledgeBaseID)
	resp, err := c.doRequest(ctx, http.MethodPut, path, request, nil)
	if err != nil {
		return nil, err
	}

	var response KnowledgeBaseResponse
	if err := parseResponse(resp, &response); err != nil {
		return nil, err
	}

	return &response.Data, nil
}

// PatchKnowledgeBaseRequest changes only the fields it carries.
// A nil field is left unchanged. Nested objects use pointers so a false or
// zero value can be sent without the other fields being cleared.
type PatchKnowledgeBaseRequest struct {
	Name             *string                     `json:"name,omitempty"`
	Description      *string                     `json:"description,omitempty"`
	EmbeddingModelID *string                     `json:"embedding_model_id,omitempty"`
	SummaryModelID   *string                     `json:"summary_model_id,omitempty"`
	StorageBackendID *string                     `json:"storage_backend_id,omitempty"`
	ChunkingConfig   *PatchChunkingConfig        `json:"chunking_config,omitempty"`
	IndexingStrategy *PatchIndexingStrategy      `json:"indexing_strategy,omitempty"`
	VLMConfig        *PatchVLMConfig             `json:"vlm_config,omitempty"`
	ASRConfig        *PatchASRConfig             `json:"asr_config,omitempty"`
	FAQConfig        *PatchFAQConfig             `json:"faq_config,omitempty"`
	AutoTagConfig    *PatchAutoTagConfig         `json:"auto_tag_config,omitempty"`
	ImageProcessing  *PatchImageProcessingConfig `json:"image_processing_config,omitempty"`
}

// PatchChunkingConfig is the partial chunking update. Nil fields stay as they are.
type PatchChunkingConfig struct {
	ChunkSize    *int      `json:"chunk_size,omitempty"`
	ChunkOverlap *int      `json:"chunk_overlap,omitempty"`
	Separators   *[]string `json:"separators,omitempty"`
}

// PatchIndexingStrategy is the partial indexing update. Nil flags stay as they are.
type PatchIndexingStrategy struct {
	VectorEnabled  *bool `json:"vector_enabled,omitempty"`
	KeywordEnabled *bool `json:"keyword_enabled,omitempty"`
	WikiEnabled    *bool `json:"wiki_enabled,omitempty"`
	GraphEnabled   *bool `json:"graph_enabled,omitempty"`
}

// PatchVLMConfig is the partial caption-model update.
type PatchVLMConfig struct {
	Enabled *bool   `json:"enabled,omitempty"`
	ModelID *string `json:"model_id,omitempty"`
}

// PatchASRConfig changes only the speech-recognition fields provided.
type PatchASRConfig struct {
	Enabled  *bool   `json:"enabled,omitempty"`
	ModelID  *string `json:"model_id,omitempty"`
	Language *string `json:"language,omitempty"`
}

// PatchFAQConfig changes only the FAQ indexing fields provided.
type PatchFAQConfig struct {
	IndexMode         *string `json:"index_mode,omitempty"`
	QuestionIndexMode *string `json:"question_index_mode,omitempty"`
}

// PatchAutoTagConfig changes only the automatic tagging fields provided.
type PatchAutoTagConfig struct {
	Enabled      *bool   `json:"enabled,omitempty"`
	ModelID      *string `json:"model_id,omitempty"`
	MaxTags      *int    `json:"max_tags,omitempty"`
	SkipIfTagged *bool   `json:"skip_if_tagged,omitempty"`
}

// PatchImageProcessingConfig changes only the image settings provided.
type PatchImageProcessingConfig struct {
	ModelID *string `json:"model_id,omitempty"`
}

// PatchKnowledgeBase changes the supplied settings and preserves omitted fields.
func (c *Client) PatchKnowledgeBase(ctx context.Context,
	knowledgeBaseID string,
	request *PatchKnowledgeBaseRequest,
) (*KnowledgeBase, error) {
	path := fmt.Sprintf("/api/v1/knowledge-bases/%s", knowledgeBaseID)
	resp, err := c.doRequest(ctx, http.MethodPatch, path, request, nil)
	if err != nil {
		return nil, err
	}

	var response KnowledgeBaseResponse
	if err := parseResponse(resp, &response); err != nil {
		return nil, err
	}

	return &response.Data, nil
}

// DeleteKnowledgeBase deletes a knowledge base
func (c *Client) DeleteKnowledgeBase(ctx context.Context, knowledgeBaseID string) error {
	path := fmt.Sprintf("/api/v1/knowledge-bases/%s", knowledgeBaseID)
	resp, err := c.doRequest(ctx, http.MethodDelete, path, nil, nil)
	if err != nil {
		return err
	}

	var response struct {
		Success bool   `json:"success"`
		Message string `json:"message,omitempty"`
	}

	return parseResponse(resp, &response)
}

// ClearKnowledgeBaseContentsResponse represents the response from clear knowledge base contents API
type ClearKnowledgeBaseContentsResponse struct {
	DeletedCount int `json:"deleted_count"`
}

// ClearKnowledgeBaseContents deletes all knowledge entries in a knowledge base (async).
// The knowledge base itself is preserved; only its contents are removed.
func (c *Client) ClearKnowledgeBaseContents(ctx context.Context, knowledgeBaseID string) (*ClearKnowledgeBaseContentsResponse, error) {
	path := fmt.Sprintf("/api/v1/knowledge-bases/%s/knowledge", knowledgeBaseID)
	resp, err := c.doRequest(ctx, http.MethodDelete, path, nil, nil)
	if err != nil {
		return nil, err
	}

	var response struct {
		Success bool                               `json:"success"`
		Message string                             `json:"message"`
		Data    ClearKnowledgeBaseContentsResponse `json:"data"`
	}

	if err := parseResponse(resp, &response); err != nil {
		return nil, err
	}

	return &response.Data, nil
}

// SearchParams represents the search parameters for hybrid search
type SearchParams struct {
	QueryText            string  `json:"query_text"`
	VectorThreshold      float64 `json:"vector_threshold"`
	KeywordThreshold     float64 `json:"keyword_threshold"`
	MatchCount           int     `json:"match_count"`
	DisableKeywordsMatch bool    `json:"disable_keywords_match"`
	DisableVectorMatch   bool    `json:"disable_vector_match"`
	// KnowledgeBaseIDs searches several knowledge bases that share one
	// embedding model; the path ID must be one of them.
	KnowledgeBaseIDs []string `json:"knowledge_base_ids,omitempty"`
	// KnowledgeIDs limits the search to these documents.
	KnowledgeIDs []string `json:"knowledge_ids,omitempty"`
	// TagIDs limits the search to these tags.
	TagIDs []string `json:"tag_ids,omitempty"`
	// Rerank, when set and enabled, reranks the candidates before cutting to
	// MatchCount. Nil keeps the raw retrieval order.
	Rerank *RerankOptions `json:"rerank,omitempty"`
}

// HybridSearch performs hybrid search.
// Pass ResourceURLOptions to receive public HTTP(S) file URLs in results.
func (c *Client) HybridSearch(
	ctx context.Context,
	knowledgeBaseID string,
	params *SearchParams,
	opts ...ResourceURLOptions,
) ([]*SearchResult, error) {
	response, err := c.HybridSearchDetailed(ctx, knowledgeBaseID, params, opts...)
	if err != nil {
		return nil, err
	}
	return response.Data, nil
}

// HybridSearchDetailed performs hybrid search and returns the whole
// response, including the rerank diagnostics in Meta.
func (c *Client) HybridSearchDetailed(
	ctx context.Context,
	knowledgeBaseID string,
	params *SearchParams,
	opts ...ResourceURLOptions,
) (*HybridSearchResponse, error) {
	path := fmt.Sprintf("/api/v1/knowledge-bases/%s/hybrid-search", knowledgeBaseID)

	queryParams := url.Values{}
	if len(opts) > 0 {
		applyResourceURLQuery(queryParams, &opts[0])
	}

	resp, err := c.doRequest(ctx, http.MethodPost, path, params, queryParams)
	if err != nil {
		return nil, err
	}

	var response HybridSearchResponse
	if err := parseResponse(resp, &response); err != nil {
		return nil, err
	}

	return &response, nil
}

// TogglePinKnowledgeBase toggles the pin status of a knowledge base.
// Server route is PUT (see internal/router/router.go); using POST silently
// 404s — the router treats unknown method on a known path as not-found,
// not 405.
func (c *Client) TogglePinKnowledgeBase(ctx context.Context, knowledgeBaseID string) (*KnowledgeBase, error) {
	path := fmt.Sprintf("/api/v1/knowledge-bases/%s/pin", knowledgeBaseID)
	resp, err := c.doRequest(ctx, http.MethodPut, path, nil, nil)
	if err != nil {
		return nil, err
	}

	var response KnowledgeBaseResponse
	if err := parseResponse(resp, &response); err != nil {
		return nil, err
	}

	return &response.Data, nil
}

// MoveTarget represents a knowledge base that can receive moved knowledge
type MoveTarget struct {
	ID          string `json:"id"`
	Name        string `json:"name"`
	Type        string `json:"type"`
	Description string `json:"description"`
}

// ListMoveTargets lists knowledge bases eligible as move targets for the given source KB
func (c *Client) ListMoveTargets(ctx context.Context, knowledgeBaseID string) ([]KnowledgeBase, error) {
	path := fmt.Sprintf("/api/v1/knowledge-bases/%s/move-targets", knowledgeBaseID)
	resp, err := c.doRequest(ctx, http.MethodGet, path, nil, nil)
	if err != nil {
		return nil, err
	}

	var response KnowledgeBaseListResponse
	if err := parseResponse(resp, &response); err != nil {
		return nil, err
	}

	return response.Data, nil
}

// CopyKnowledgeBase copies a knowledge base asynchronously and returns task info
func (c *Client) CopyKnowledgeBase(ctx context.Context, request *CopyKnowledgeBaseRequest) (*CopyKnowledgeBaseResponse, error) {
	path := "/api/v1/knowledge-bases/copy"

	resp, err := c.doRequest(ctx, http.MethodPost, path, request, nil)
	if err != nil {
		return nil, err
	}

	var response struct {
		Success bool                      `json:"success"`
		Data    CopyKnowledgeBaseResponse `json:"data"`
	}

	if err := parseResponse(resp, &response); err != nil {
		return nil, err
	}

	return &response.Data, nil
}

// DuplicateKnowledgeBase creates a settings-only duplicate and returns the new knowledge base info.
func (c *Client) DuplicateKnowledgeBase(
	ctx context.Context,
	knowledgeBaseID string,
) (*DuplicateKnowledgeBaseResponse, error) {
	path := fmt.Sprintf("/api/v1/knowledge-bases/%s/duplicate", knowledgeBaseID)

	resp, err := c.doRequest(ctx, http.MethodPost, path, nil, nil)
	if err != nil {
		return nil, err
	}

	var response struct {
		Success bool                           `json:"success"`
		Data    DuplicateKnowledgeBaseResponse `json:"data"`
	}

	if err := parseResponse(resp, &response); err != nil {
		return nil, err
	}

	return &response.Data, nil
}

// GetKBCloneProgress gets the progress of a knowledge base clone task
func (c *Client) GetKBCloneProgress(ctx context.Context, taskID string) (*KBCloneProgress, error) {
	path := fmt.Sprintf("/api/v1/knowledge-bases/copy/progress/%s", taskID)

	resp, err := c.doRequest(ctx, http.MethodGet, path, nil, nil)
	if err != nil {
		return nil, err
	}

	var response struct {
		Success bool            `json:"success"`
		Data    KBCloneProgress `json:"data"`
	}

	if err := parseResponse(resp, &response); err != nil {
		return nil, err
	}

	return &response.Data, nil
}