# WeKnora | Tencent | Knowledge Base | Instagit

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

GitHub Stars: 22.6k

Repository: https://github.com/tencent/WeKnora

---

## Articles

### [How to Install and Use the WeKnora TypeScript Client SDK](/tencent/WeKnora/how-to-install-and-use-the-weknora-typescript-client-sdk)

Install and use the WeKnora TypeScript client SDK with npm or Yarn. Instantiate WeKnoraClient for typed requests to the WeKnora REST API.

- Tags: how-to-guide
- Published: 2026-09-13

### [Which Messaging Platforms Does WeKnora Integrate With? Complete Platform Support Guide](/tencent/WeKnora/which-messaging-platforms-does-weknora-integrate-with)

Discover which messaging platforms WeKnora integrates with. Explore support for WeCom, Slack, Telegram, WeChat, and more in this comprehensive guide.

- Tags: api-reference
- Published: 2026-09-13

### [Authentication and Authorization Mechanisms in WeKnora: A Complete Technical Guide](/tencent/WeKnora/what-are-the-authentication-and-authorization-mechanisms-in-weknora)

Explore WeKnora's robust authentication and authorization mechanisms. Learn how JWT, API keys, OIDC, and RBAC ensure secure, fine-grained access control for your applications.

- Tags: how-to-guide
- Published: 2026-09-13

### [How WeKnora Implements Event Streaming and Tracing: A Deep Dive into the Go Source Code](/tencent/WeKnora/how-does-weknora-implement-event-streaming-and-tracing)

Explore how WeKnora implements event streaming and tracing with its dual-subsystem architecture. Learn about the StreamManager interface and TracingContext for enhanced observability.

- Tags: deep-dive
- Published: 2026-09-13

### [How WeKnora Handles Multi-Instance Storage Backends: Architecture and Implementation](/tencent/WeKnora/how-does-weknora-handle-multi-instance-storage-backends)

Discover how WeKnora manages multi-instance storage backends using UUIDs, StorageBackendID, and runtime resolution via FileServiceResolver. Learn its architecture and implementation.

- Tags: architecture
- Published: 2026-09-13

### [WeKnora Sandbox Backends: Cube vs E2B Configuration Guide](/tencent/WeKnora/what-are-the-different-sandbox-backends-supported-by-weknora)

Explore WeKnora sandbox backends, comparing Cube vs E2B configuration. Choose between stateful Cube Sandbox or stateless E2B Sandbox for your WeKnora setup.

- Tags: how-to-guide
- Published: 2026-09-13

### [How WeKnora Manages Task Queues and Worker Pools: Architecture and Implementation](/tencent/WeKnora/how-does-weknora-manage-its-task-queue-and-worker-pools)

Discover how WeKnora manages task queues and worker pools using Redis and PostgreSQL for efficient and atomic document processing. Explore its architecture and implementation.

- Tags: architecture
- Published: 2026-09-13

### [What the WeKnora modelcontext Package Does: Request-Scoped Model Handling](/tencent/WeKnora/what-is-the-modelcontext-package-responsible-for-in-weknora)

Learn how the WeKnora modelcontext package handles request-scoped models isolating temporary handles encoding LLM payloads and enforcing tool policies for secure and efficient operations.

- Tags: internals
- Published: 2026-09-13

### [How the ReAct Loop Works in WeKnora's `internal/agent` Package](/tencent/WeKnora/how-does-the-react-loop-work-in-weknoras-internal-agent-package)

Discover how the ReAct loop in WeKnora's internal agent package orchestrates LLM calls for iterative reasoning and action. Learn about its executeLoop and runReActIteration functions to understand its thought process and tool e...

- Tags: internals
- Published: 2026-09-13

### [What the internal/application Package Handles in WeKnora: Business Logic and Service Layer](/tencent/WeKnora/what-does-the-internal-application-package-handle-in-weknora)

Discover how the WeKnora internal/application package manages business logic, orchestrating wiki lifecycles, content ingestion, skill execution, and vector search operations.

- Tags: internals
- Published: 2026-09-13

### [WeKnora Internal Handler Package: Controller Layer Architecture and Implementation](/tencent/WeKnora/what-is-the-purpose-of-the-internal-handler-package-in-weknora)

Explore the WeKnora internal handler package as the controller layer. Learn how it exposes HTTP endpoints, enforces RBAC, and manages task queuing for business logic.

- Tags: architecture
- Published: 2026-09-13

### [How to Configure Observability with Langfuse in WeKnora](/tencent/WeKnora/how-to-configure-observability-with-langfuse-in-weknora)

Learn how to configure Langfuse observability in Tencent WeKnora. Easily track LLM generations with automatic activation and minimal runtime overhead. Get started today.

- Tags: how-to-guide
- Published: 2026-09-13

### [How to Enable Wiki and Knowledge Graph Features in WeKnora: A Complete Setup Guide](/tencent/WeKnora/how-to-enable-wiki-and-knowledge-graph-features-in-weknora)

Learn to enable Wiki and Knowledge Graph features in WeKnora with our complete setup guide. Configure Neo4j and knowledge base settings for seamless integration.

- Tags: how-to-guide
- Published: 2026-09-13

### [How to Integrate Browser Skills into WeKnora: Complete Setup and Usage Guide](/tencent/WeKnora/how-to-integrate-browser-skills-into-weknora)

Learn to integrate browser skills into WeKnora with our setup and usage guide. Execute native browser actions via WebSocket CDP commands without local sandboxing. Explore Tencent/BrowserSkill today.

- Tags: how-to-guide
- Published: 2026-09-13

### [How to Enable Embed Subdomain and Secure Mode in WeKnora](/tencent/WeKnora/how-to-enable-embed-subdomain-and-secure-mode-in-weknora)

Learn how to enable embed subdomain and secure mode in WeKnora by configuring your reverse proxy, creating an embed channel, and switching to short-lived session tokens.

- Tags: how-to-guide
- Published: 2026-09-13

### [How to Configure Chunking Strategies in WeKnora: A Complete API and SDK Guide](/tencent/WeKnora/how-to-configure-chunking-strategies-in-weknora)

Learn to configure WeKnora chunking strategies via the API and SDK. Choose from heading, heuristic, auto, or recursive tiers for optimal knowledge base organization.

- Tags: how-to-guide
- Published: 2026-09-13

### [How to Create and Manage Agent Skills in WeKnora: A Complete Developer Guide](/tencent/WeKnora/how-to-create-and-manage-agent-skills-in-weknora)

Learn how to create and manage agent skills in WeKnora. Discover reusable skill bundles, REST endpoints, UI management, and sandbox integration for your developer projects.

- Tags: how-to-guide
- Published: 2026-09-13

### [How to Develop a Custom Data Source Connector for WeKnora: A Complete Implementation Guide](/tencent/WeKnora/how-to-develop-a-custom-data-source-connector-for-weknora)

Learn how to develop a custom data source connector for WeKnora. Implement the datasource Connector interface and register it to integrate your data seamlessly.

- Tags: how-to-guide
- Published: 2026-09-13

### [How to Add a New Web Search Provider to WeKnora: Complete Integration Guide](/tencent/WeKnora/how-to-add-a-new-web-search-provider-to-weknora)

Learn how to add a new web search provider to WeKnora. This guide covers backend registration, API exposure, frontend mapping, and utility configuration for seamless integration.

- Tags: how-to-guide
- Published: 2026-09-13

### [How to Integrate Different LLM Providers with WeKnora: A Complete Guide](/tencent/WeKnora/how-to-integrate-different-llm-providers-with-weknora)

Easily integrate diverse LLM providers like OpenAI, Azure, and HuggingFace with WeKnora. Learn how to switch between models seamlessly without code changes using our complete guide.

- Tags: how-to-guide
- Published: 2026-09-13

### [How to Customize the Vector Store in WeKnora: Configuration and Deployment Guide](/tencent/WeKnora/how-to-customize-the-vector-store-in-weknora)

Customize your WeKnora vector store by configuring engine specific connections and index settings or using environment variables for hybrid retrieval. Deploy easily with this guide.

- Tags: how-to-guide
- Published: 2026-09-13

### [How WeKnora's Architecture Is Layered: Three-Process Core and Go Backend Tiers](/tencent/WeKnora/how-is-weknoras-architecture-layered)

Explore WeKnora's layered architecture. Discover its three-process core including Vue3 frontend, Go backend tiers (Router, Handler, Service, Repository, Infra), and Python reader. Learn how Uber-dig powers dependency injection.

- Tags: architecture
- Published: 2026-09-13

### [The Three WeKnora Reasoning Modes: Quick‑Answer, Agent, and Smart‑Reasoning Explained](/tencent/WeKnora/what-are-the-three-reasoning-modes-in-weknora)

Explore WeKnora's three reasoning modes: Quick-Answer, Agent, and Smart-Reasoning. Understand how each mode processes queries for efficient retrieval and autonomous tool execution.

- Tags: deep-dive
- Published: 2026-09-13

### [How to Run WeKnora Locally: Docker and Development Setup Guide](/tencent/WeKnora/how-to-run-weknora-locally)

Run WeKnora locally with Docker Compose for full containerization or use fast development mode with hot-reload for efficient coding. Get your WeKnora setup today.

- Tags: how-to-guide
- Published: 2026-09-13

### [How `resource_urls=public` Short‑Circuits the Second Authenticated Proxy Call for IM/Embed Clients](/tencent/WeKnora/resource_urls-public-short-circuit-proxy-call-im-embed-clients)

Discover how resource_urls=public short-circuits the second authenticated proxy call for IM/embed clients. Fetch assets in a single HTTP request by rewriting resource:// handles.

- Tags: internals
- Published: 2026-09-12

### [How WeKnora Automatic Database Migrations Run on Upgrade and Handle Partial Failures](/tencent/WeKnora/automatic-database-migration-upgrade-partial-failure-handling)

Learn how WeKnora automatic database migrations run on upgrade. Discover how it detects dirty states and handles partial failures with configurable auto-recovery for seamless upgrades.

- Tags: how-to-guide
- Published: 2026-09-12

### [How Session-Persistent Skill Sandboxes Enforce Per-Tenant Network Policy in WeKnora](/tencent/WeKnora/session-persistent-skill-sandboxes-network-policy-enforcement)

Learn how WeKnora's session-persistent skill sandboxes enforce per-tenant network policy by storing firewall rules and injecting restrictions into container runtimes.

- Tags: internals
- Published: 2026-09-12

### [How OIDC ID-Token JWKS Verification Integrates with Login and Tenant Provisioning in WeKnora](/tencent/WeKnora/oidc-id-token-jwks-verification-login-tenant-provisioning)

Learn how WeKnora uses OIDC ID-token JWKS verification to securely integrate login and tenant provisioning. Ensure only validated identities get organizational access.

- Tags: how-to-guide
- Published: 2026-09-12

### [How WeKnora Implements AES-256-GCM At-Rest Encryption with Key Rotation for Credentials](/tencent/WeKnora/aes-256-gcm-encryption-key-rotation-credential-handling)

Learn how WeKnora uses AES-256-GCM at-rest encryption with key rotation to protect credentials. Discover its secure key management and re-encryption process.

- Tags: how-to-guide
- Published: 2026-09-12

### [Queue and DLQ Design for Scaling Wiki Ingest to 40k Documents per Knowledge Base in WeKnora](/tencent/WeKnora/queue-dlq-design-wiki-ingest-scaling-40k-documents)

Discover the robust queue and DLQ design in Tencent WeKnora that scales Wiki ingest to 40k documents per knowledge base using Redis and Asynq for efficient processing and error handling.

- Tags: architecture
- Published: 2026-09-12

### [WeKnora Hybrid Retrieval Pipeline: Combining BM25, Dense Vectors, GraphRAG, and Parent-Child Chunking](/tencent/WeKnora/hybrid-retrieval-pipeline-bm25-dense-graphrag-chunking)

Explore WeKnora's hybrid retrieval pipeline that combines BM25 dense vectors GraphRAG and parent-child chunking for precise contextually grounded answers.

- Tags: deep-dive
- Published: 2026-09-12

### [How WeKnora Tunes and Rebuilds the HNSW‑Accelerated pgvector Index for 1024‑Dim Embeddings](/tencent/WeKnora/hnsw-pgvector-index-tuning-rebuilding-1024-dim-embeddings)

Learn how WeKnora tunes and rebuilds the HNSW pgvector index for 1024-dim embeddings. Optimize migration and query performance with recommended settings for faster vector search.

- Tags: performance
- Published: 2026-09-12

### [How the Document Parsing Trace Timeline Reconstructs Langfuse-Style Span Trees in WeKnora](/tencent/WeKnora/document-parsing-trace-timeline-reconstruct-langfuse-span-tree)

Learn how WeKnora reconstructs Langfuse-style span trees by persisting processing metadata to PostgreSQL and querying parent-child relationships for its Vue frontend timeline.

- Tags: internals
- Published: 2026-09-12

### [How the WeKnora System-Admin Dashboard Displays Failed Tasks and Enables Manual Retries](/tencent/WeKnora/system-admin-task-queue-dashboard-failed-tasks-manual-retry)

Learn how the WeKnora system-admin dashboard visualizes failed tasks and enables manual retries through its Vue interface and runtime-task inspector. Retry jobs with ease.

- Tags: how-to-guide
- Published: 2026-09-12

### [How WeKnora Embed Widget Secure-Mode Token Exchange and Rate Limiting Works](/tencent/WeKnora/website-embed-widget-secure-mode-token-exchange-rate-limiting)

Discover how WeKnora embed widget secure-mode handles token exchange and rate limiting to protect your application. Learn about publish and session tokens.

- Tags: how-to-guide
- Published: 2026-09-12

### [How MCP Server Mid-Conversation OAuth2 Flow Works with Embed and MCP Sessions](/tencent/WeKnora/mcp-server-oauth2-flow-embed-mcp-session-scopes)

Explore the MCP server's mid-conversation OAuth2 flow. Learn how it handles blocking interactive clients while allowing embed sessions to continue seamlessly. Understand the WeKnora approach to token negotiation.

- Tags: deep-dive
- Published: 2026-09-12

### [How to Wire a Custom LLM Provider Using the OpenAI-Compatible API in WeKnora](/tencent/WeKnora/wire-custom-llm-provider-openai-compatible-api)

Effortlessly integrate custom LLM providers with WeKnora using the OpenAI-compatible API. Configure base_url or use the LiteLLM adapter for seamless customization.

- Tags: how-to-guide
- Published: 2026-09-12

### [Vector Store Driver Abstraction in WeKnora: How Tencent Built Interchangeable Backends for Milvus and Qdrant](/tencent/WeKnora/vector-store-driver-abstraction-interchangeable-backends-milvus-qdrant)

Discover how Tencent's WeKnora uses a factory-registry pattern to abstract vector stores, enabling interchangeable backends like Milvus and Qdrant without code changes.

- Tags: architecture
- Published: 2026-09-12

### [How to Implement a New Data Source Connector in WeKnora: A Complete Guide Following Feishu, GitLab, and Notion Patterns](/tencent/WeKnora/implement-new-data-source-connector-feishu-gitlab-notion-layout)

Implement a new data source connector in WeKnora by creating a package, implementing the Connector interface, and registering metadata. Follow Feishu, GitLab, and Notion patterns for seamless integration.

- Tags: how-to-guide
- Published: 2026-09-12

### [How to Add a New Web Search Provider to WeKnora: Exa or Metaso Integration Guide](/tencent/WeKnora/add-new-web-search-provider-exa-metaso-pattern)

Easily add Exa or Metaso web search providers to WeKnora. Learn how to implement and register new search providers for enhanced retrieval capabilities.

- Tags: how-to-guide
- Published: 2026-09-12

### [How WeKnora Manages Per-Knowledge-Base Model Selection and Thinking-Mode Override](/tencent/WeKnora/per-knowledge-base-model-selection-thinking-mode-override)

Discover how WeKnora manages per knowledge base model selection and allows API override of thinking modes. Learn to customize your LLM reasoning for Tencent WeKnora.

- Tags: internals
- Published: 2026-09-12

### [WeKnora Principal Model for API Keys and MCP/Embed Session Isolation](/tencent/WeKnora/principal-model-api-keys-mcp-embed-session-isolation)

Discover WeKnora's principal model for API key and session isolation. It uses typed identifiers to separate human and machine authentication, securing your API keys, tokens, and embed sessions.

- Tags: architecture
- Published: 2026-09-12

### [How Scoped API Keys with Capability-Level Grants and KB Restrictions Function in WeKnora](/tencent/WeKnora/scoped-api-keys-capability-grants-kb-restrictions-functionality)

Learn how WeKnora's scoped API keys with capability grants and KB restrictions secure machine-to-machine interactions by authorizing specific actions on designated datasets.

- Tags: how-to-guide
- Published: 2026-09-12

### [How Multi-Instance Storage Backends Are Bound Per Workspace and Knowledge Base in WeKnora](/tencent/WeKnora/multi-instance-storage-backends-binding-workspace-knowledge-base)

Discover how WeKnora binds multi-instance storage backends per workspace and knowledge base. Learn about hierarchical storage binding and tenant isolation.

- Tags: internals
- Published: 2026-09-12

### [How the WeKnora Runtime Task-Queue Dashboard Schedule Works with Per-Stage Worker-Pool Governance](/tencent/WeKnora/runtime-task-queue-dashboard-scheduling-worker-pool-governance)

Discover how WeKnora's runtime task-queue dashboard leverages Redis Asynq servers and worker-pool governance for real-time scheduling, per-stage control, and elastic capacity sharing.

- Tags: internals
- Published: 2026-09-12

### [How Cross-Session Long-Term Memory Is Auto-Extracted and Stored in WeKnora](/tencent/WeKnora/cross-session-long-term-memory-auto-extraction-storage)

Discover how WeKnora automatically extracts and stores cross-session long-term memory using LLMs and a background queue worker for efficient retrieval.

- Tags: deep-dive
- Published: 2026-09-12

### [WeKnora Wiki Mode Auto-Generation Pipeline: Architecture and Revision History](/tencent/WeKnora/wiki-mode-auto-generation-pipeline-architecture-revision-history)

Explore the WeKnora Wiki Mode auto-generation pipeline architecture. Learn how it transforms documents into linked Markdown pages and manages revision history efficiently.

- Tags: architecture
- Published: 2026-09-12

### [How the Tencent/WeKnora ReAct Agent Handles Retrieval, MCP Tools, and Skill Sandboxes Each Turn](/tencent/WeKnora/how-react-agent-orchestrates-retrieval-mcp-tools-skill-sandboxes)

Discover how the Tencent WeKnox ReAct Agent uses a think-analyze-act-observe cycle to manage retrieval, MCP tools, and skill sandboxes for dynamic knowledge access and isolated code execution each turn.

- Tags: deep-dive
- Published: 2026-09-12