"Creating a Custom Local Knowledge Base" matching MCP servers:

Delegates mechanical text-to-text tasks to a local LLM to save Claude subscription quota, reading files server-side so large content never enters Claude's context.

Provides offline translation across 200+ languages using a local NLLB-200 model, with no data leaving your machine.

Adds a single MCP tool that lets Codex request local, read-only draft transformations of user-provided text, using a local Ollama model for translation, summarization, rewriting, extraction, and classification.

Integrates local language models (like Qwen3-8B) with MCP clients, providing tools for chat, code analysis, text generation, translation, and content summarization using your own hardware.

A local-first multilingual dictionary, thesaurus, translation, lexical-relations, semantic-neighbour, and English wordplay server for the Model Context Protocol (MCP). It operates fully offline with read-only query access and no data mutation while running.

Finds new UI translation keys in your codebase, dedup-checks them against a Blocks/UILM tenant, and generates the JSON the Blocks portal expects for import.

Enables AI agents to conjugate verbs in 14 European languages via a hosted Rust API, requiring only a free API key.

Enables agents to speak Thai and English audio through local speakers, manage a playback queue, save speech to files, translate text into Thai, and OCR PDFs and images.

A language expression coach MCP server that provides native-sounding translations with cultural context, tone notes, and audio pronunciation for over 13 languages.

Enables translation, language detection, and language list retrieval via MCP tools, acting as a thin wrapper over a self-hosted LibreTranslate instance.

prevents llm from overload, freeze and oscillation

Provides AI models with tools to proofread and format Chinese text before delivery, covering typos, punctuation, casing, spacing, and platform-specific character limits. It runs as a zero-dependency MCP service so the model checks and fixes its own Chinese writing automatically.

Translation QA for i18n locale files. Deterministic validators — missing keys, placeholder drift, collapsed plurals, glossary breaches — that need no API key, plus a review tool that hands translation pairs and judging criteria back to the calling agent, so meaning-level review runs on the model your client already has. Translation with your own LLM key is optional. JSON, Flutter .arb, Apple .

A Docker-based MCP server wrapping the Smartling CLI, enabling natural language management of translation projects, files, and glossaries.

MCP server for HumanPen that lets AI agents work on real documents (.docx, .pptx, .pdf) — humanizing content to lower AI-detection scores, converting citations, condensing, and translating while preserving formatting, tables, images, and citations.

Provides local anonymization of Czech legal documents by replacing sensitive entities with pseudonyms to ensure privacy during LLM interactions. It allows users to safely process documents like contracts and judgments by keeping original data offline and facilitating local deanonymization.

Delegates replaceable grunt work (boilerplate, formatting, translation, long-document summarizing) from a premium agent to cheap models behind a local LiteLLM proxy, auto-routing each task by type. Delegated calls run in a separate process, so the subscription session and the API credentials never share an environment.

Enables OCR of local images through a public endpoint and writes the extracted text to a specified file, optimized for Tibetan, Sanskrit, and Devanagari input with optional transliteration.

Provides AI agents with local Dutch spelling and word validation using OpenTaal/hunspell, plus detailed word information from woordenlijst.org. It enables checking full texts or individual words and retrieving linguistic details for Dutch writing tasks.

A Model Context Protocol server that enables users to manage translations for their applications through SimpleLocalize's localization platform via natural language prompts in Cursor.