🌎 DialectOS

The first Model Context Protocol server built specifically for Spanish dialects.

DialectOS is an open-source Spanish dialect translation server that runs as an MCP (Model Context Protocol) tool and CLI. It translates English and other languages into 25 regional Spanish variants — Mexican, Argentinian, Colombian, Puerto Rican, and more — while preserving markdown structure, enforcing glossary terms, and applying adversarial quality gates that catch semantic drift before it reaches users.

Translate, detect, and adapt content across 25 regional Spanish variants while preserving markdown structure, code comments, and locale file formatting.

📖 Documentation · 🚀 Quick Start · 🛠️ MCP Tools · Agent Skill · 📦 Packages · 🤝 Contributing · 📋 Roadmap

Public Discovery

DialectOS is a Spanish localization and dialect QA system for AI agents, documentation teams, app developers, and support organizations. It provides MCP tools, CLI workflows, glossary enforcement, locale-file validation, and adversarial quality gates for regional Spanish variants.

AI discovery: llms.txt provides a compact project summary for AI assistants and search crawlers.

Best-fit searches: Spanish dialect translation MCP server, Spanish localization QA, Model Context Protocol translation tool, i18n validation CLI, regional Spanish translator, glossary enforcement, AI localization audit, Spanish launch certification.

Agent Skill

DialectOS includes a public agent skill at skills/dialectos/SKILL.md. Use $dialectos in compatible agent hosts when you want an agent to choose the right MCP or CLI workflow for regional Spanish translation, markdown preservation, locale-file validation, glossary enforcement, register checks, and launch-readiness QA.

Spanish Launch Certification

DialectOS is available as a paid Spanish localization launch audit. We certify your Spanish docs, app strings, support macros, or locale files across target dialects and deliver an MQM-aligned launch-readiness report.

• Beta pilot: $500

• Scope: up to 10,000 source words and 5 target dialects

• Deliverables: certification report, issue list, severity table, recommended fixes, launch decision

• Sample report: audits/sample-customer-report.md

• Offer details: docs/spanish-launch-certification.md

✨ What makes DialectOS different?

🎯 Why this exists

"We shipped a product to Mexico using our Spain Spanish translations. Users thought we were being intentionally rude."

Spanish is not one language — it's 25 regional variants with different vocabulary, formality levels, slang, and grammatical preferences. Existing translation tools treat Spanish as a monolith.

DialectOS solves this by:

• Understanding regional differences (es-MX vs es-ES vs es-AR vs es-CO...)

• Preserving technical document structure during translation

• Providing glossary enforcement for consistent terminology

• Adding semantic context, dialect grammar profiles, quality contracts, and quality gates that catch drift before it reaches users

• Running as an MCP server so AI assistants can translate natively

🚀 Quick Start

Note: DialectOS packages are not yet published to npm. Setup requires cloning the repo and building from source (~2–5 minutes first time).

Full-app browser demo

The browser demo is no longer a fake/static rule replacer. It calls a local DialectOS backend, and that backend calls the configured provider stack.

LLM_API_URL="http://127.0.0.1:1234/v1/chat/completions" \
LLM_API_FORMAT="openai" \
LLM_MODEL="your-local-model-name" \
LLM_ALLOW_LOCAL=1 \
pnpm demo

Open http://127.0.0.1:8080.

For the beginner container walkthrough, see docs/full-app-demo.md.

Local MCP setup

After pnpm build, add to your Claude Desktop, Cursor, or any MCP client:

{
  "mcpServers": {
    "dialectos": {
      "command": "node",
      "args": ["packages/mcp/dist/index.js"],
      "comment": "Local development setup — see README for clone and build instructions.",
      "env": {
        "LLM_API_URL": "https://your-llm-gateway/v1/chat/completions",
        "LLM_MODEL": "your-dialect-capable-model",
        "LLM_API_KEY": "your-key-if-required",
        "LLM_API_FORMAT": "openai",
        "ALLOWED_LOCALE_DIRS": "/path/to/locales"
      }
    }
  }
}

Recommended certified models

For v0.3.0, the recommended default cloud model is glm-4.5-air through the Z.ai international Anthropic-compatible endpoint. It passed basic, expanded adversarial, and long-document certification. Use glm-5.1 when you want the higher-confidence/premium option, and qwen3.5-9b via LM Studio for local/offline certification.

export LLM_API_URL="https://api.z.ai/api/anthropic/v1/messages"
export LLM_MODEL="glm-4.5-air"
export LLM_API_FORMAT="anthropic"
export LLM_API_KEY="..."

LM Studio local model testing

Start LM Studio's local server, then point DialectOS at any downloaded local model. LLM_API_FORMAT=lmstudio uses LM Studio's native REST API and loads the model just-in-time when needed.

LM_STUDIO_URL="http://127.0.0.1:1234" \
LLM_MODEL="publisher/model-key-or-api-identifier" \
LLM_API_FORMAT="lmstudio" \
pnpm dialect:eval -- --live --provider=llm --out=/tmp/dialectos-lmstudio-eval

Incremental provider certification

Use dialect:certify for long local-model or cloud-provider runs. It writes events.jsonl, progress.json, and an incrementally updated results.json after every sample, with per-sample timeout protection.

LM_STUDIO_URL="http://127.0.0.1:1234" \
LLM_MODEL="qwen3.5-9b" \
LLM_API_FORMAT="lmstudio" \
pnpm dialect:certify -- --live --provider=llm --sample-timeout-ms=300000 --out=/tmp/dialectos-certify

Adversarial dialect certification

Use dialect:certify:adversarial to run paraphrase, dialect-collision, taboo-copy, placeholder, register, and repeatability traps. It wraps dialect:certify and writes a failure-matrix.md plus aggregate repeatability results.

pnpm dialect:certify:adversarial -- --live --provider=llm --repeat=2 --sample-timeout-ms=300000 --out=/tmp/dialectos-adversarial

Long-document certification

Use dialect:certify:documents to certify README/API-doc/locale JSON flows, not just sentence fixtures. It checks markdown structure, placeholders, URLs, code fences, API tables, and locale JSON outputs.

pnpm dialect:certify:documents -- --live --provider=llm --dialects=es-MX,es-PA,es-PR --out=/tmp/dialectos-doc-cert

Customer-facing certification report

Use dialect:report to turn certification artifacts into a customer-facing Markdown deliverable for paid launch audits.

pnpm dialect:report -- --input=audits/release-candidate-2026-04-22/model-matrix.json --out=customer-report.md --customer="Acme SaaS" --product="Spanish launch"

CLI install

# Install the v0.3.0 CLI tarball from the GitHub Release
pnpm add -g https://github.com/KyaniteLabs/DialectOS/releases/download/v0.3.0/dialectos-cli-0.3.0.tgz

# Or use a local source checkout
pnpm install --frozen-lockfile
pnpm build

# Translate to Mexican Spanish
dialectos translate "Hello world" --dialect es-MX

# Translate a README preserving structure
dialectos translate-readme README.md --dialect es-AR --output README.ar.md

# Validate an existing translation
dialectos validate --source "Click the button" --translated "Haz clic en el botón" --dialect es-MX

# Validate translation files
dialectos validate --source-file en.json --translated-file es-MX.json --dialect es-MX --format json

# View translation corpus statistics
dialectos corpus stats

# Run dialect quality benchmark
dialectos benchmark run --dialects es-MX,es-AR,es-ES

# Generate glossary suggestions from corrections
dialectos glossary suggest --min-occurrences 3

# Compare two glossary versions
dialectos glossary diff glossary-v1.json glossary-v2.json

# Detect missing i18n keys
dialectos i18n detect-missing ./locales/en.json ./locales/es.json

# List all supported dialects
dialectos dialects list

From source

git clone https://github.com/KyaniteLabs/DialectOS.git
cd DialectOS
pnpm install
pnpm build
pnpm test        # 662+ tests passing

🛠️ MCP Tools

Markdown Translation (4 tools)

i18n Operations (6 tools)

Translation (7 tools)

📦 Packages

662+ tests across 7 packages plus docs contracts, demo-server contracts, and static-hardening checks

🛡️ Security & Quality

DialectOS has undergone adversarial security hardening:

• 18 CVEs resolved via dependency overrides

• SSRF protection on all provider endpoints

• Circuit breaker with half-open probe locks

• Atomic checkpoint writes with schema versioning

• HTML injection detection in translated output

• Semantic drift scoring — catches "looks valid but meaning changed"

• Provider capability negotiation — validates language support before API calls

• Chaos harness for deterministic resilience testing

See SECURITY.md for details.

🎨 Supported Dialects

...and 12 more. Full list via dialectos dialects list.

🏗️ Architecture

┌─────────────────────────────────────────────────────────────┐
│                        MCP Client                            │
│              (Claude Desktop / Cursor / etc.)                │
└──────────────────────┬──────────────────────────────────────┘
                       │ stdio
┌──────────────────────▼──────────────────────────────────────┐
│                   @dialectos/mcp                               │
│              17 tools • JSON-RPC over stdio                  │
└──────────────────────┬──────────────────────────────────────┘
                       │
┌──────────────────────▼──────────────────────────────────────┐
│                   @dialectos/cli                               │
│   translate • validate • corpus • benchmark • glossary     │
│   ├─ Policy profiles (strict/balanced/permissive)           │
│   ├─ Quality gates (token/glossary/structure/semantic)      │
│   ├─ Translation corpus + auto-glossary                     │
│   └─ Checkpoint resumption + telemetry                      │
└──────────────────────┬──────────────────────────────────────┘
                       │
┌──────────────────────▼──────────────────────────────────────┐
│                @dialectos/providers                            │
│   ┌─────────┐  ┌─────────────────┐  ┌─────────────────┐    │
│   │   LLM   │  │     DeepL       │  │ Libre/MyMemory │   │
│   │ Primary │  │ Paid fallback   │  │ Generic fallback│    │
│   └─────────┘  └─────────────────┘  └─────────────────┘    │
│        │                │                    │              │
│        └────────────────┴────────────────────┘              │
│              Circuit Breaker + Rate Limiter                  │
└─────────────────────────────────────────────────────────────┘

📊 Quality Gates

Every translation passes through 4 quality dimensions:

Quality Score = tokenIntegrity×25% + glossaryFidelity×30% + structureIntegrity×20% + semanticSimilarity×25%

❓ FAQ

What is DialectOS? DialectOS is an open-source translation engine for Spanish regional dialects. It runs as an MCP server (for AI assistants like Claude) and a CLI tool for developers.

How is DialectOS different from Google Translate? Google Translate treats Spanish as one language. DialectOS understands 25 regional variants, preserves markdown structure, enforces glossaries, and applies quality gates that catch errors before they reach users.

What are Spanish dialects? Spanish varies significantly by country. Mexican Spanish uses "carro" for car; Spain uses "coche"; Argentina uses "auto". DialectOS handles these differences automatically.

Does DialectOS work with ChatGPT / Claude? Yes. DialectOS is an MCP server, so Claude Desktop, Cursor, Windsurf, and other MCP clients can use its 17 translation tools natively.

Is DialectOS free? Yes. DialectOS v0.3.0 is released under Apache-2.0. See LICENSE for details.

What is MCP? Model Context Protocol is an open standard that lets AI assistants use external tools. DialectOS exposes 17 translation tools through MCP so AI agents can translate natively.

Can I use DialectOS for commercial projects? Yes. Apache-2.0 allows commercial use, modification, and redistribution subject to the license terms. See LICENSE for details.

How accurate is the translation? DialectOS applies 4 quality gates (token integrity, glossary fidelity, structure integrity, semantic similarity) and adversarial tests. Automated tests verify correctness across dialects.

🏷️ Badges

Add this badge to your project if you use DialectOS for translation:

[![Translated with DialectOS](https://img.shields.io/badge/translated%20with-DialectOS-d89b2b)](https://github.com/KyaniteLabs/DialectOS)

⚡ GitHub Action

Validate Spanish translations in CI on every pull request:

- uses: KyaniteLabs/DialectOS/action  # Planned — version pinning unavailable until first release
  with:
    dialect: es-MX
    source-dir: locales/en
    target-patterns: 'locales/es-MX/*.json'
    fail-on-blocking: true

Multi-dialect matrix:

strategy:
  matrix:
    dialect: [es-ES, es-MX, es-AR, es-CO]
steps:
  - uses: KyaniteLabs/DialectOS/action  # Planned — version pinning unavailable until first release
    with:
      dialect: ${{ matrix.dialect }}
      fail-on-blocking: true

See docs/github-action.md for full configuration options.

🤝 Contributing

We welcome contributors! See CONTRIBUTING.md for:

• Setting up your development environment

• Running the test suite

• Submitting pull requests

• Code style guidelines

Good first issues are tagged with good first issue — perfect for newcomers.

📋 Roadmap

See ROADMAP.md for upcoming features including:

• Portuguese dialect support (pt-BR, pt-PT)

• Real-time collaborative translation

• Custom provider plugins

• OpenAI-compatible, Anthropic-compatible, and LM Studio local gateways via LLM_API_URL/LM_STUDIO_URL + LLM_MODEL + LLM_API_FORMAT

• VS Code extension

📄 License

Apache-2.0 — see LICENSE for details.

Made with ❤️ by KyaniteLabs and contributors.

Star ⭐ this repo if it helps your project!

Part of KyaniteLabs

More from KyaniteLabs. Related projects:

• mcp-video — guardrailed video-editing MCP server for AI agents

• Epoch — time-estimation MCP server (PERT) for AI agents

• checkyourself — local-first production-readiness checks for AI-built code

→ More at kyanitelabs.tech