# minisearch | Victor Nogueira | Knowledge Base | Instagit

Minimalist web-searching platform with an AI assistant that runs directly from your browser. Uses WebLLM, Wllama and SearXNG. Demo: https://felladrin-minisearch.hf.space

GitHub Stars: 547

Repository: https://github.com/felladrin/minisearch

---

## Articles

### [How to Optimize Cache Performance and Tune Cache Metrics in MiniSearch](/felladrin/minisearch/how-to-optimize-cache-performance-tune-cache-metrics)

Optimize MiniSearch cache performance with TTL and entry limits. Monitor hit rates and tune pruning intervals for efficient query speed and storage.

- Tags: performance
- Published: 2026-03-01

### [MiniSearch Verification Token System: Secure Search with verifiedTokens and handleTokenVerification](/felladrin/minisearch/explain-verification-token-system-verifiedtokens-handletokenverification)

Learn how MiniSearch secures its search endpoint with a verification token system. Explore verifiedTokens and handleTokenVerification for robust protection and rate limiting.

- Tags: how-to-guide
- Published: 2026-03-01

### [How to Add Custom Text Generation Backends to MiniSearch: A Complete Developer Guide](/felladrin/minisearch/how-to-add-custom-text-generation-backends-architecture)

Unlock powerful custom text generation backends for MiniSearch. This guide details how developers can extend the architecture with new inference types and dispatch logic to integrate seamlessly.

- Tags: how-to-guide
- Published: 2026-03-01

### [How Cross-Origin Resource Sharing (CORS) Configuration Works in MiniSearch Server Hooks](/felladrin/minisearch/how-cross-origin-resource-sharing-cors-configuration-server-hooks)

Learn how MiniSearch implements CORS configuration using custom Vite server hooks and cross-origin isolation headers for enhanced security and SharedArrayBuffer support.

- Tags: how-to-guide
- Published: 2026-03-01

### [How to Deploy MiniSearch Using Docker for Development and Production Environments](/felladrin/minisearch/how-to-deploy-minisearch-using-docker-development-production)

Deploy MiniSearch with Docker. Follow steps for development and production environments using docker-compose for efficient container management and rapid iteration.

- Tags: how-to-guide
- Published: 2026-03-01

### [How the Circuit Breaker Pattern Is Implemented in the Minisearch Server Utilities](/felladrin/minisearch/how-circuit-breaker-pattern-implemented-server-utilities)

Discover how the circuit breaker pattern is implemented in Minisearch server utilities. Learn about its three-state model to prevent cascading failures in async operations.

- Tags: internals
- Published: 2026-03-01

### [MiniSearch Server Environment Variables: Complete Configuration Reference](/felladrin/minisearch/what-server-environment-variables-available-for-configuration)

Explore over 18 MiniSearch server environment variables for complete configuration of security, networking AI routing, and model behavior. Get full details now.

- Tags: api-reference
- Published: 2026-03-01

### [Search History Cleanup Logic in MiniSearch: Retention Days, Max Entries, and Auto-Cleanup Explained](/felladrin/minisearch/search-history-cleanup-retention-days-max-entries-auto-cleanup)

Understand MiniSearch search history cleanup logic. Learn how retention days, max entries, and auto cleanup discard old or excessive entries while protecting pinned items.

- Tags: internals
- Published: 2026-03-01

### [How to Integrate MiniSearch with External OpenAI-Compatible APIs: Base URL, API Key, and Model Configuration](/felladrin/minisearch/how-to-integrate-with-external-openai-compatible-apis)

Integrate MiniSearch with external OpenAI-compatible APIs. Easily configure Base URL, API Key, and Model for seamless inference using simple environment variables.

- Tags: how-to-guide
- Published: 2026-03-01

### [WebLLM Model Download and Caching in the Browser: A Complete Guide](/felladrin/minisearch/webllm-model-download-caching-browser-environment)

Learn how WebLLM downloads and caches models in your browser using IndexedDB for efficient offline inference. Discover reusable model assets.

- Tags: how-to-guide
- Published: 2026-03-01

### [How to Configure Inference Parameters (Temperature, Top‑P, Min‑P, and Penalties) in MiniSearch](/felladrin/minisearch/how-to-configure-inference-parameters-temperature-topp-minp-penalties)

Learn to configure MiniSearch inference parameters like temperature, topP, minP, and penalties. Control generation settings via UI, API, or Docker for optimal results.

- Tags: how-to-guide
- Published: 2026-03-01

### [How the Pub/Sub State Management System Enables Real-Time Updates in MiniSearch](/felladrin/minisearch/how-pub-sub-state-management-system-works)

Discover how MiniSearch uses the Pub/Sub state management system with create-pubsub for efficient real-time UI updates in React components. Learn about localStorage persistence and throttled AI stream handling.

- Tags: internals
- Published: 2026-03-01

### [Debugging Issues Using the MiniSearch Log Entries System: A Complete Guide](/felladrin/minisearch/debugging-issues-with-log-entries-system)

Master debugging with MiniSearch's in-browser log entries system. Learn best practices for privacy-safe debugging without external servers. A complete guide for felladrin/minisearch.

- Tags: how-to-guide
- Published: 2026-03-01

### [How Image Search Functionality Works in MiniSearch: A Complete Technical Guide](/felladrin/minisearch/how-image-search-functionality-implemented)

Explore MiniSearch image search functionality. Discover its type-safe client-server pipeline querying SearXNG, reranking, fetching thumbnails, and caching in IndexedDB.

- Tags: technical-guide
- Published: 2026-03-01

### [How the Search Results Reranker Service Operates in MiniSearch](/felladrin/minisearch/how-search-results-reranker-service-works)

Discover how MiniSearch reranks search results using a local LLM for enhanced relevance. Learn about its Jina model integration and statistical filtering.

- Tags: internals
- Published: 2026-03-01

### [How Minisearch Implements Search Token Hash Authentication](/felladrin/minisearch/how-search-token-hash-system-works-authentication)

Discover how Minisearch implements search token hash authentication using Argon2id for secure API access. Learn about client-side hashing and server-side verification.

- Tags: internals
- Published: 2026-03-01

### [How to Implement Password Protection Using the ACCESS_KEYS Environment Variable in MiniSearch](/felladrin/minisearch/how-to-implement-password-protection-access-keys-env-variable)

Implement password protection in MiniSearch using the ACCESS_KEYS environment variable. Secure API endpoints with Argon2 validation and prompt users for authentication.

- Tags: how-to-guide
- Published: 2026-03-01

### [How to Customize the System Prompt and Citation Formatting in MiniSearch](/felladrin/minisearch/how-to-customize-system-prompt-citation-formatting)

Learn to customize MiniSearch system prompt and citation formatting. Edit client/modules/settings.ts and client/modules/textGenerationUtilities.ts to control LLM search result appearance.

- Tags: how-to-guide
- Published: 2026-03-01

### [How MiniSearch Manages Long Conversations: Conversation Memory, Rolling Summaries, and Token Budgets](/felladrin/minisearch/conversation-memory-summary-system-rolling-summary-token-budget)

Discover how MiniSearch manages long conversations with rolling summaries and token budgets. Learn how it preserves context up to 800 tokens within its active chat window.

- Tags: deep-dive
- Published: 2026-03-01

### [How to Set Up and Configure the Internal OpenAI-Compatible API in MiniSearch](/felladrin/minisearch/how-to-set-up-internal-openai-compatible-api)

Learn how to set up and configure the internal OpenAI-compatible API in MiniSearch. Easily proxy requests to your self-hosted LLM via the inference endpoint.

- Tags: how-to-guide
- Published: 2026-03-01

### [Browser-Based AI Model Loading and WebGPU Integration in MiniSearch](/felladrin/minisearch/browser-based-ai-model-loading-webgpu-integration)

Explore browser-based AI model loading with WebGPU integration in MiniSearch. Learn how this architecture enables efficient LLM execution directly in your browser.

- Tags: deep-dive
- Published: 2026-03-01

### [How to Configure and Use AI Text Generation Backends in MiniSearch](/felladrin/minisearch/how-to-configure-and-use-ai-text-generation-backends-webllm-wllama-openai-horde-internal-api)

Learn to configure and use AI text generation backends in MiniSearch. Explore WebLLM, Wllama, OpenAI, AI Horde, and internal APIs with this guide. Optimize your search experience.

- Tags: how-to-guide
- Published: 2026-03-01

### [How MiniSearch Search Caching Works with IndexedDB, TTL, and Max Entry Limits](/felladrin/minisearch/how-does-minisearch-search-caching-mechanism-work-indexeddb-ttl-max-entries)

Discover how MiniSearch search caching uses IndexedDB, TTL, and max entries to efficiently store and manage query results, ensuring fast performance and controlled storage.

- Tags: internals
- Published: 2026-03-01