PHPackages netresearch/nr-llm - PHPackages - PHPackages [Skip to content](#main-content)[PHPackages](/)[Directory](/)[Categories](/categories)[Trending](/trending)[Leaderboard](/leaderboard)[Changelog](/changelog)[Analyze](/analyze)[Collections](/collections)[Log in](/login)[Sign up](/register) 1. [Directory](/) 2. / 3. [Localization & i18n](/categories/localization) 4. / 5. netresearch/nr-llm ActiveTypo3-cms-extension[Localization & i18n](/categories/localization) netresearch/nr-llm ================== Shared AI/LLM foundation for TYPO3 — centralized provider management, encrypted API keys, and ready-to-use services for chat, translation, vision, and embeddings v0.5.0(2mo ago)23.9k—0%1[3 issues](https://github.com/netresearch/t3x-nr-llm/issues)1GPL-2.0-or-laterPHPPHP ^8.2CI passing Since Jan 11Pushed 1mo ago1 watchersCompare [ Source](https://github.com/netresearch/t3x-nr-llm)[ Packagist](https://packagist.org/packages/netresearch/nr-llm)[ Docs](https://github.com/netresearch/t3x-nr-llm)[ RSS](/packages/netresearch-nr-llm/feed)WikiDiscussions main Synced 1mo ago READMEChangelog (10)Dependencies (19)Versions (32)Used By (1) nr-llm — The Shared AI Foundation for TYPO3 =========================================== [](#nr-llm--the-shared-ai-foundation-for-typo3) [![CI](https://github.com/netresearch/t3x-nr-llm/actions/workflows/ci.yml/badge.svg)](https://github.com/netresearch/t3x-nr-llm/actions/workflows/ci.yml)[![codecov](https://camo.githubusercontent.com/ef89c9355a8934b29b1d71e9822e442becfa991f47419d4d32c045231ed27528/68747470733a2f2f636f6465636f762e696f2f67682f6e657472657365617263682f7433782d6e722d6c6c6d2f67726170682f62616467652e737667)](https://codecov.io/gh/netresearch/t3x-nr-llm)[![Documentation](https://github.com/netresearch/t3x-nr-llm/actions/workflows/docs.yml/badge.svg)](https://github.com/netresearch/t3x-nr-llm/actions/workflows/docs.yml)[![OpenSSF Scorecard](https://camo.githubusercontent.com/2cce612ba1aea0034c5d46d9eb9b8eef2dc78add7ebb57765327afc262a26948/68747470733a2f2f6170692e736563757269747973636f726563617264732e6465762f70726f6a656374732f6769746875622e636f6d2f6e657472657365617263682f7433782d6e722d6c6c6d2f6261646765)](https://securityscorecards.dev/viewer/?uri=github.com/netresearch/t3x-nr-llm)[![OpenSSF Best Practices](https://camo.githubusercontent.com/4c0d499d678bd62201efdc50427078585c19180a3e6f328166e44db6c690d8db/68747470733a2f2f7777772e626573747072616374696365732e6465762f70726f6a656374732f31313639372f6261646765)](https://www.bestpractices.dev/projects/11697)[![PHPStan](https://camo.githubusercontent.com/d18b9a987aa81e64470a11caecf72caa66597c9ebd6b307bd1c2cb7a752b0dff/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f5048505374616e2d6c6576656c25323031302d627269676874677265656e2e737667)](https://phpstan.org/)[![PHP 8.2+](https://camo.githubusercontent.com/0f16581d1180dbfd4c0e13166ec1267d4ad2f2fab8281ea6d6b284cf5c65d921/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f5048502d382e322532422d626c75652e737667)](https://www.php.net/)[![TYPO3 v13.4+](https://camo.githubusercontent.com/55f8fb237b0ea7b977760a9c819113ec71df597862190a8cba75d6059d38c7c0/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f5459504f332d7631332e342532422d6f72616e67652e737667)](https://typo3.org/)[![License: GPL v2](https://camo.githubusercontent.com/6b639abd353f330706524560924b761ba87ba1895911daf38f4b71b9347bb94f/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f4c6963656e73652d47504c5f76322d626c75652e737667)](https://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html)[![Latest Release](https://camo.githubusercontent.com/613d85c07bbfb53f9ba0e7f96f967c6ea712bdf3d24f0e92765a133abee9fb94/68747470733a2f2f696d672e736869656c64732e696f2f6769746875622f762f72656c656173652f6e657472657365617263682f7433782d6e722d6c6c6d)](https://github.com/netresearch/t3x-nr-llm/releases)[![Contributor Covenant](https://camo.githubusercontent.com/29a162993634c8b29ac484c10c05c2db4be4e6ea7a916cf5c130711e91d26364/68747470733a2f2f696d672e736869656c64732e696f2f62616467652f436f6e7472696275746f72253230436f76656e616e742d322e302d3462616161612e737667)](CODE_OF_CONDUCT.md)[![SLSA 3](https://camo.githubusercontent.com/dc294f15fb5f1c96307863a1e96860310be940504e7ee370cee94bf4400cbac9/68747470733a2f2f736c73612e6465762f696d616765732f67682d62616467652d6c6576656c332e737667)](https://slsa.dev) **One LLM setup. Every extension. Full admin control.** nr-llm is shared infrastructure for AI in TYPO3 — like the caching framework, but for language models. Administrators configure providers once; every AI-powered extension on the site uses them automatically. The Problem ----------- [](#the-problem) Every TYPO3 extension that wants AI capabilities today has to: - Build its own provider integration (HTTP calls, auth, error handling, streaming) - Store API keys in its own way (often plaintext in extension settings) - Create its own backend configuration UI - Leave administrators with no central overview of AI usage or costs When a site runs three AI extensions, that's three separate API key configurations, three places to check when something breaks, and no way to switch providers globally. The Solution ------------ [](#the-solution) nr-llm provides the missing shared layer: ``` ┌─────────────────────────────────────────────────┐ │ Your Extension │ Cowriter │ SEO Assistant │ │ (3 lines of DI) │ │ │ └────────┬─────────┴─────┬─────┴────────┬────────┘ │ │ │ ┌────────▼───────────────▼──────────────▼────────┐ │ nr-llm Service Layer │ │ Chat · Translation · Vision · Embeddings │ │ Streaming · Tool Calling · Caching │ ├─────────────────────────────────────────────────┤ │ Provider Abstraction Layer │ │ OpenAI · Anthropic · Gemini · Ollama · … │ ├─────────────────────────────────────────────────┤ │ Admin Tools > LLM (Backend Module) │ │ Encrypted keys · Usage tracking · Setup wizard │ └─────────────────────────────────────────────────┘ ``` --- For Extension Developers ------------------------ [](#for-extension-developers) **Add AI to your TYPO3 extension in 5 minutes** — no API key handling, no HTTP client code, no provider-specific logic. ### 1. Require the package [](#1-require-the-package) ``` composer require netresearch/nr-llm ``` ### 2. Inject the service you need [](#2-inject-the-service-you-need) ``` use Netresearch\NrLlm\Service\LlmServiceManagerInterface; class MyController { public function __construct( private readonly LlmServiceManagerInterface $llm, ) {} public function summarizeAction(string $text): string { return $this->llm->complete("Summarize: {$text}")->content; } } ``` That's it. Provider selection, API keys, caching, error handling — all managed by nr-llm. ### What you get for free [](#what-you-get-for-free) CapabilityWithout nr-llmWith nr-llmProvider switchingRewrite HTTP callsChange one admin settingAPI key storage`$GLOBALS` or plaintextEncrypted (sodium / nr-vault)Response cachingBuild your ownBuilt-in, TYPO3 caching frameworkStreaming (SSE)Implement per provider`foreach ($llm->streamChat($msg) as $chunk)`Error handlingParse each provider's errorsTyped exceptions with provider contextMultiple providersN × integration effortOne interface, all providers### Available services [](#available-services) ``` // Chat & completion $response = $llm->chat($messages); $response = $llm->complete('Explain TYPO3 content elements'); // Translation $result = $translationService->translate('Hello world', 'de'); // Vision / image analysis $altText = $visionService->generateAltText($imageUrl); // Embeddings & similarity $vector = $embeddingService->embed('semantic search query'); // Streaming foreach ($llm->streamChat($messages) as $chunk) { echo $chunk; } // Tool/function calling $response = $llm->chatWithTools($messages, $toolDefinitions); ``` ### Why not call the OpenAI API directly? [](#why-not-call-the-openai-api-directly) You can — but then your extension only works with OpenAI. Your users can't switch to Anthropic, use a local Ollama instance, or route through Azure. And every extension on the site manages its own API keys, its own error handling, its own caching. nr-llm solves this once for the entire TYPO3 ecosystem. > **[Developer Guide](Documentation/Developer/Index.rst)** — Full API reference, custom provider registration, typed options, response objects > > **[Integration Guide](Documentation/Developer/IntegrationGuide.rst)** — Step-by-step tutorial for building your extension on nr-llm --- For TYPO3 Administrators ------------------------ [](#for-typo3-administrators) ### One dashboard for all AI [](#one-dashboard-for-all-ai) The **Admin Tools > LLM** backend module gives you full control: - **Providers** — Register API connections (OpenAI, Anthropic, Gemini, Ollama, …) - **Models** — Define which models are available and their capabilities - **Configurations** — Create use-case presets (temperature, system prompts, token limits) - **Tasks** — Define reusable prompt templates for editors ### Security by default [](#security-by-default) - **Encrypted API keys** — All keys stored with sodium\_crypto\_secretbox (XSalsa20-Poly1305), optionally via [nr-vault](https://github.com/netresearch/t3x-nr-vault) envelope encryption - **Admin-only access** — Backend module restricted to administrators - **No plaintext secrets** — Keys never stored or logged in plain text ### Setup in 2 minutes [](#setup-in-2-minutes) The **Setup Wizard** auto-detects your provider type from the endpoint URL, discovers available models, and generates a ready-to-use configuration. Paste your API key and go. ### AI-powered wizards [](#ai-powered-wizards) - **Task Wizard** — Describe what you need in plain language; the AI generates a complete task with configuration, system prompt, and model recommendation in one step - **Configuration Wizard** — AI-assisted configuration generation with system prompts, parameters, and model selection - **Model Discovery** — "Fetch Models" button on model fields queries the provider API and auto-fills capabilities and pricing ### Supported providers [](#supported-providers) ProviderAdapterCapabilitiesOpenAI`openai`Chat, Embeddings, Vision, Streaming, ToolsAnthropic Claude`anthropic`Chat, Vision, Streaming, ToolsGoogle Gemini`gemini`Chat, Embeddings, Vision, Streaming, ToolsOllama`ollama`Chat, Embeddings, Streaming (local, no API key)OpenRouter`openrouter`Chat, Embeddings, Vision, Streaming, ToolsMistral`mistral`Chat, Embeddings, StreamingGroq`groq`Chat, Streaming (fast inference)Azure OpenAI`azure_openai`Same as OpenAIAny OpenAI-compatible`custom`Varies (vLLM, LocalAI, LiteLLM, …)--- For Agencies & Solution Architects -------------------------------------- [](#for-agencies--solution-architects) - **Reduce integration effort** — AI capabilities across client projects without per-project plumbing - **No vendor lock-in** — Switch from OpenAI to Anthropic (or a local model) without code changes - **Compliance-friendly** — Encrypted keys, admin-only access, SBOM and SLSA provenance on every release - **Local-first option** — Ollama support means AI features work without sending data to external APIs - **Production-proven** — Powers [t3x-cowriter](https://github.com/netresearch/t3x-cowriter), the CKEditor 5 AI writing assistant for TYPO3 --- Built on nr-llm --------------- [](#built-on-nr-llm) ExtensionWhat it doesnr-llm services used[t3x-cowriter](https://github.com/netresearch/t3x-cowriter)AI writing assistant in CKEditor 5Chat, Streaming, Translation, Tasks*Building on nr-llm? [Open a PR](https://github.com/netresearch/t3x-nr-llm/pulls) to add your extension here.* --- Architecture ------------ [](#architecture) The three-tier configuration hierarchy separates concerns cleanly: ``` ┌─────────────────────────────────────────────────────────┐ │ CONFIGURATION (Use-Case Specific) │ │ "blog-summarizer", "product-description", "translator" │ │ → system_prompt, temperature, max_tokens │ └───────────────────────────┬─────────────────────────────┘ │ references ┌───────────────────────────▼─────────────────────────────┐ │ MODEL (Available Models) │ │ "gpt-5.3-instant", "claude-sonnet-4-6", "llama-70b" │ │ → model_id, capabilities, pricing │ └───────────────────────────┬─────────────────────────────┘ │ references ┌───────────────────────────▼─────────────────────────────┐ │ PROVIDER (API Connections) │ │ "openai-prod", "openai-dev", "local-ollama" │ │ → endpoint, api_key (encrypted), adapter_type │ └─────────────────────────────────────────────────────────┘ ``` **Benefits:** - Multiple API keys per provider type (prod/dev/backup) - Custom endpoints (Azure OpenAI, Ollama, vLLM, local models) - Reusable model definitions across configurations - Clear separation of concerns --- Requirements ------------ [](#requirements) - PHP 8.2+ - TYPO3 v13.4+ or v14.x - PSR-18 HTTP Client (e.g., guzzlehttp/guzzle) Installation ------------ [](#installation) ``` composer require netresearch/nr-llm ``` Then activate in **Admin Tools > Extensions** and run **Admin Tools > LLM > Setup Wizard**. --- Documentation ------------- [](#documentation) - **[Introduction](Documentation/Introduction/Index.rst)** — Overview and use cases - **[Configuration](Documentation/Configuration/Index.rst)** — Backend module setup - **[Developer Guide](Documentation/Developer/Index.rst)** — API reference, services, custom providers - **[Integration Guide](Documentation/Developer/IntegrationGuide.rst)** — Build your extension on nr-llm - **[Architecture](Documentation/Architecture/Index.rst)** — Design decisions and ADRs --- Supply Chain Security --------------------- [](#supply-chain-security) This project implements supply chain security best practices: ### SLSA Level 3 Provenance [](#slsa-level-3-provenance) All releases include [SLSA Level 3](https://slsa.dev) build provenance attestations, providing: - **Non-falsifiable provenance**: Cryptographically signed attestations - **Isolated build environment**: Builds run in GitHub Actions with no user-controlled steps - **Source integrity**: Provenance links artifacts to exact source commit ### Verifying Release Artifacts [](#verifying-release-artifacts) ``` # Install slsa-verifier go install github.com/slsa-framework/slsa-verifier/v2/cli/slsa-verifier@latest # Download release artifacts gh release download v1.0.0 -R netresearch/t3x-nr-llm # Verify SLSA provenance slsa-verifier verify-artifact nr_llm-1.0.0.zip \ --provenance-path multiple.intoto.jsonl \ --source-uri github.com/netresearch/t3x-nr-llm \ --source-tag v1.0.0 ``` ### Verifying Signatures (Cosign) [](#verifying-signatures-cosign) All release artifacts are signed using [Sigstore Cosign](https://www.sigstore.dev/) keyless signing: ``` # Install cosign go install github.com/sigstore/cosign/v2/cmd/cosign@latest # Verify signature cosign verify-blob nr_llm-1.0.0.zip \ --signature nr_llm-1.0.0.zip.sig \ --certificate nr_llm-1.0.0.zip.pem \ --certificate-identity-regexp 'https://github.com/netresearch/t3x-nr-llm/' \ --certificate-oidc-issuer 'https://token.actions.githubusercontent.com' ``` ### Software Bill of Materials (SBOM) [](#software-bill-of-materials-sbom) Each release includes SBOMs in both [SPDX](https://spdx.dev/) and [CycloneDX](https://cyclonedx.org/) formats: - `nr_llm-X.Y.Z.sbom.spdx.json` - SPDX format - `nr_llm-X.Y.Z.sbom.cdx.json` - CycloneDX format ### Checksums [](#checksums) SHA256 checksums for all artifacts are provided in `checksums.txt`, which is also signed. License ------- [](#license) GPL-2.0-or-later Author ------ [](#author) Netresearch DTT GmbH ### Health Score 43 — FairBetter than 91% of packages Maintenance69 Regular maintenance activity Popularity29 Limited adoption so far Community16 Small or concentrated contributor base Maturity48 Maturing project, gaining track record Bus Factor1 Top contributor holds 88.4% of commits — single point of failure How is this calculated?**Maintenance (25%)** — Last commit recency, latest release date, and issue-to-star ratio. Uses a 2-year decay window. **Popularity (30%)** — Total and monthly downloads, GitHub stars, and forks. Logarithmic scaling prevents top-heavy scores. **Community (15%)** — Contributors, dependents, forks, watchers, and maintainers. Measures real ecosystem engagement. **Maturity (30%)** — Project age, version count, PHP version support, and release stability. ### Release Activity Cadence Every ~2 days Total 22 Last Release 62d ago PHP version history (2 changes)v0.1.0PHP ^8.5 v0.2.0PHP ^8.2 ### Community Maintainers ![](https://www.gravatar.com/avatar/acffee6a64e18f21593794b335dd8786001148f7df89fd8372a54d3dd09d91a4?d=identicon)[netresearch](/maintainers/netresearch) --- Top Contributors [![CybotTM](https://avatars.githubusercontent.com/u/326348?v=4)](https://github.com/CybotTM "CybotTM (502 commits)")[![dependabot[bot]](https://avatars.githubusercontent.com/in/29110?v=4)](https://github.com/dependabot[bot] "dependabot[bot] (27 commits)")[![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?v=4)](https://github.com/github-actions[bot] "github-actions[bot] (16 commits)")[![renovate[bot]](https://avatars.githubusercontent.com/in/2740?v=4)](https://github.com/renovate[bot] "renovate[bot] (16 commits)")[![just-tobi](https://avatars.githubusercontent.com/u/5242689?v=4)](https://github.com/just-tobi "just-tobi (7 commits)") --- Tags aianthropicclaudeembeddingsgeminigptllmollamaopenaiphpprovider-abstractionstreamingtypo3typo3-extensiontranslationaistreamingopenaiextensiontypo3Geminiclaudechatbotllmollamaembeddingsprovider-abstraction ### Code Quality Static AnalysisPHPStan, Rector Code StylePHP CS Fixer Type Coverage Yes ### Embed Badge ![Health badge](/badges/netresearch-nr-llm/health.svg) ``` [![Health](https://phpackages.com/badges/netresearch-nr-llm/health.svg)](https://phpackages.com/packages/netresearch-nr-llm) ``` ### Alternatives [cognesy/instructor-php The complete AI toolkit for PHP: unified LLM API, structured outputs, agents, and coding agent control 310107.9k1](/packages/cognesy-instructor-php)[theodo-group/llphant LLPhant is a library to help you build Generative AI applications. 1.5k311.5k5](/packages/theodo-group-llphant)[symfony/ai-platform PHP library for interacting with AI platform provider. 51927.7k134](/packages/symfony-ai-platform)[vizra/vizra-adk Vizra Agent Development Kit - A comprehensive Laravel package for building intelligent AI agents. 29026.1k](/packages/vizra-vizra-adk)[sbsaga/toon 🧠 TOON for Laravel — a compact, human-readable, and token-efficient data format for AI prompts & LLM contexts. Perfect for ChatGPT, Gemini, Claude, Mistral, and OpenAI integrations (JSON ⇄ TOON). 6115.6k](/packages/sbsaga-toon) PHPackages © 2026 [Directory](/)[Categories](/categories)[Trending](/trending)[Changelog](/changelog)[Analyze](/analyze)