## 1. Project Overview & Quickstart (score-spec/GOVERNANCE) # GOVERNANCE Open-source repository score-spec/GOVERNANCE ### Repository Details - **Repository:** [score-spec/GOVERNANCE](https://github.com/score-spec/GOVERNANCE) - **Primary Language:** Code *Note: High-volume repository documentation is actively indexed and synchronized by YakaAI.* ## 2. Official Technical Reference & Guides (score-spec/docs) ## File: README.md [](https://github.com/score-spec/docs/actions/workflows/hugo.yml) [](https://github.com/score-spec/docs/issues?q=is%3Aissue+is%3Aopen+label%3A%22good+first+issue%22) ## Docs This section covers how to build the documentation site with Hugo. For information on Score's style guide see the [Style guide](styles/style-guide.md). ### Running the website locally #### `yarn` We use `yarn` (for MacOS: `brew install yarn`) as a package manager to ensure the versions of Hugo and associated tools are up to date. All commands that execute those tools are prefixed with `yarn run` or are aliased in the `package.json` file. To install the packages: ``` yarn install ``` To run the server locally: ``` yarn build OR yarn run hugo server ``` And then view it at . #### `score-compose` You can [install `score-compose`](https://docs.score.dev/docs/score-implementation/score-compose/installation/) and then build and run this hugo website as a container: ``` score-compose init --no-sample score-compose generate score.yaml --build main=. docker compose up --build -d curl $(score-compose resources get-outputs dns.default#score-docs.dns --format '{{ .host }}:8080/docs/') ``` ### Deployment This site is currently deployed through Github Pages with a CNAME directing `docs.score.dev` to `score-spec.github.io`. The DNS configuration is managed by Humanitec. ### Format docs The following section covers how to format and lint prose. ### Formatters and linters This project uses the [dprint](https://dprint.dev/overview/) to format documentation. dprint is a command line application that automatically formats code. Use `dprint` to format your documentation. ```bash yarn fmt ``` Example output. ```bash $ yarn fmt Formatted 1 file. ✨ Done in 0.13s. ``` This project uses the [Vale](https://vale.sh) with a [Vale-compatible implementation of the Google Developer Documentation Style Guide](https://github.com/errata-ai/Google). On MacOS, install it through `brew install vale`. For example, to lint a document run: ```bash vale sync vale styles/style-guide.md ``` Example output. ```bash styles/style-guide.md 40:71 error Did you really mean Vale.Spelling 'inclusivity'? ✖ 1 error, 0 warnings and 0 suggestions in 1 file. ``` To lint all documents, run `yarn lint`. ## Autogenerated content ### Score example hub This site uses content from external Git repositories to create and continuously update the "Score example hub" pages. The external repos are pulled in to the `/gen/external-content` folder. The commands for the initial integration of the repos are: ```bash git remote add -f -t main --no-tags examples https://github.com/score-spec/examples.git git remote add -f -t main --no-tags community-provisioners https://github.com/score-spec/community-provisioners.git git remote add -f -t main --no-tags score-compose https://github.com/score-spec/score-compose.git git remote add -f -t main --no-tags score-k8s https://github.com/score-spec/score-k8s.git git remote add -f -t main --no-tags patch-templates https://github.com/score-spec/community-patchers.git git read-tree --prefix=gen/external-content/score/specification -u examples/main:specification git read-tree --prefix=gen/external-content/resource-provisioners/default/score-compose -u score-compose/main:internal/command git read-tree --prefix=gen/external-content/score/resources/default-provisioners -u examples/main:resources git read-tree --prefix=gen/external-content/score/resources/community-provisioners -u community-provisioners/main git read-tree --prefix=gen/external-content/resource-provisioners/default/score-k8s -u score-k8s/main:internal/provisioners/default git read-tree --prefix=gen/external-content/resource-provisioners/community -u community-provisioners/main git read-tree --prefix=gen/external-content/patch-templates -u patch-templates/main git add gen/external-content git commit -s -S -m "Integrating external content" ``` These commands will not have to be repeated unless re-creating the repo integration, or moving the source location. In that case, remove the `remote`, delete the local contents, repeat these commands targeting the new location, and update the generation scripts. To generate the example hub pages based on the external content, execute this command: ```bash yarn gen-example-pages ``` ### Pulling new content To pull the current content from the remote example library repo, execute this command: ```bash yarn gen-get-external-content ``` Refer to the [package.json`](./package.json) to see the actual implementation of this command. ## Troubleshooting documentation site builds This section covers common build issues with Hugo. ### Failed to resolve output format `print` When building the server, you may receive the following error message. ```bash Error: from config: failed to resolve output format "print" from site config error Command failed with exit code 255. info Visit https://yarnpkg.com/en/docs/cli/run for documentation about this command. ``` To resolve this issue, delete the temporary Hugo cache directory. By default, [`-cacheDir`](https://gohugo.io/commands/hugo_server/) is stored at `$TMPDIR/hugo_cache`. ```bash rm -rf $TMPDIR/hugo_cache ``` ### `package-lock.json` locked files When running `yarn`, you may receive the following error message. ```bash warning package-lock.json found. Your project contains lock files generated by tools other than Yarn. It is advised not to mix package managers in order to avoid resolution inconsistencies caused by unsynchronized lock files. To clear this warning, remove package-lock.json. ``` To resolve this issue, delete the `package-lock.json` file. ```bash rm package-lock.json ``` You can find our documentation at [docs.score.dev](https://docs.score.dev/docs/). ### Contribute If you have a suggestion to improve our documentation, please fork the repo and create a pull request. You can also open an issue with the tag `enhancement`. 1. Fork the Project. 2. Create your Feature Branch. `git checkout -b feature/feature-name` 3. Commit your Changes. `git commit -s -m "Add some amazing documentation enhancement"` 4. Push to the Branch. `git push origin feature/feature-name` 5. Open a Pull Request. Read [CONTRIBUTING](CONTRIBUTING.md) for more information. ### Roadmap See [Roadmap](roadmap.md). You can [submit an idea](https://github.com/score-spec/spec/issues/new) anytime. ### License [](https://opensource.org/licenses/Apache-2.0) ### Code of conduct [](CODE_OF_CONDUCT.md) --- ## File: content/en/docs/community/_index.md --- title: "Community" linkTitle: "Community" weight: 10 Victor_Hugo: "true" Focus_Keyword: "Join the Score community" url: /docs/community aliases: - /docs/contribute - /docs/talks - /docs/presentations - /docs/cncf - /docs/samples - /docs/examples --- ## Join the Score community Score is open-source and a [Cloud Native Computing Foundation (CNCF) Sandbox project](https://www.cncf.io/projects/score/)! Connect with fellow Score developers on our [Slack](https://cloud-native.slack.com/archives/C07DN0D1UCW) channel on the CNCF Slack ([slack.cncf.io](https://slack.cncf.io/)) or start find your way to [contribute to Score](https://github.com/score-spec/spec/blob/main/CONTRIBUTING.md)! Got something to say or ask about Score? [We're all ears](https://github.com/score-spec/spec/discussions)! We’d love to discuss your ideas on where the project should go next or how we can improve. We host [regular community meetings](https://github.com/score-spec/spec?tab=readme-ov-file#-get-in-touch) to discuss updates, share ideas, and collaborate. ## Watch talks about Score - [KubeCon IN 2026 - The Death of the YAML-Engineer: Engineering "Invisible" Platforms With Crossplane and Score](https://sched.co/2IW4r) [[video](https://youtu.be/JEWSboHLuQc)] - [Open Source Summit NA 2026 - GitOps Gone Wild: Hardening Delivery Pipelines for the AI Era](https://sched.co/2JQuJ) [[video](https://youtu.be/hFVKJBTc7o4?list=PLbzoR-pLrL6p2URzlq8xlNtBhPspd2xpi)] - [Open Source Summit NA 2026 - From Apps To Infrastructure: A Cloud Native First Approach](https://sched.co/2JQws) [[video](https://youtu.be/SXBUv_I_Bfc?list=PLbzoR-pLrL6p2URzlq8xlNtBhPspd2xpi)] - [KubeCon EU 2026 - Unifying Inner & Outer Loops To Bridge the Gaps Between Devs & Ops With Microcks + Score](https://sched.co/2CVxb) [[video](https://youtu.be/Y_jfBWbBTZQ?list=PLj6h78yzYM2MXCOWSN9CqqID6OOvF7wxL)] - [Cloud Native Rejekts NA 2025 - Make your Developer's Pains go Away, with the Right Level of Abstraction for your Platform](https://youtu.be/CLCK2lBpWa8) - [LF Live Webinar: From CI to Kubernetes Catalog: A Practical Guide to vCluster and GitOps](https://youtu.be/U4AgmEo3oV8) - [KubeCon EU 2025 - Score Lightning Talk](https://sched.co/1tcwp) [[video](https://youtu.be/Nq_PgPKZHsc?list=PLj6h78yzYM2MP0QhYFK8HOb8UqgbIkLMc)] - [KubeCon EU 2025 - Platform Engineering Loves Security: Shift Down To Your Platform, Not Left To Your Developers!](https://sched.co/1txGE) [[video](https://youtu.be/Es3DBj2UgIE?list=PLj6h78yzYM2MP0QhYFK8HOb8UqgbIkLMc)] - [KubeCon EU 2025 - Dapr + Score: Mixing the Perfect Cocktail for an Enhanced Developer Experience](https://sched.co/1txGi) [[video](https://youtu.be/-fGztPUuD8k?list=PLj6h78yzYM2MP0QhYFK8HOb8UqgbIkLMc)] - [You Choose!, Ch. 05, Ep. 05 - Specialized Templating - Feat. Porter, Werf, Radius, Score, PipeCD](https://youtu.be/TEZVeWsirsw) - [Cloud Native Rejekts NA 2024 - You Can Score It! Shift Down to the Platform. Do Not Shift Left to the Developers](https://youtu.be/qdk5mLEnrNo) - [ChatLoopBackOff - Episode 23 (Score)](https://www.youtube.com/watch?v=BRiZ0t6MYNo&list=PLj6h78yzYM2PnyOsrsCbR_kqjCKfPObHK&index=22) - [Open at Microsoft - Improve the Developers Experience from your Platform with Score](https://learn.microsoft.com/en-us/shows/open-at-microsoft/improve-the-developers-experience-from-your-platform-with-score) - [PlatformCon 2024 - Dapr + Score: Crafting and improving your developers experience](https://platformcon.com/talks/dapr-score-crafting-and-improving-your-developers-experience) - [PlatformCon 2024 - Bridging the gap between local and remote environments in your platform](https://platformcon.com/talks/bridging-the-gap-between-local-and-remote-environments-in-your-platform) - [CNCF WG Platforms with Score](https://www.youtube.com/watch?v=P6otOxdOKDk&t=1260s) - [Dapr Community Call - Dapr and Score](https://www.youtube.com/watch?v=4gT0Y6QxinU&t=141s) - [InfraCloud - Deep Dive into Workload Specification with Score](https://www.infracloud.io/webinars/deep-dive-into-workload-specification-with-score/) ## Read external blog posts about Score - [Anatomy of a score.yaml: How Resolution Actually Works](https://rootcause.hashnode.dev/anatomy-of-a-score-yaml-how-resolution-actually-works) - [Score + Docker Compose to deploy your local LLM models](https://medium.com/google-cloud/score-docker-compose-to-deploy-your-local-llm-models-10aff89686ce) - [Unifying the Inner & Outer Loops to Bridge the Gaps between Devs & Ops with Containers + Microcks + Score](https://itnext.io/unifying-inner-outer-loops-to-bridge-the-gaps-between-devs-ops-with-containers-microcks-d28603342f4b) - [Generate your Backstage software catalog files with Score](https://medium.com/@mabenoit/generate-your-backstage-software-catalog-files-with-score-b62aa33e8ecc) - [Deploy Backstage with Score](https://medium.com/@mabenoit/deploy-backstage-with-score-45bb2d7c2d90) - [Stop Building Platforms Nobody Uses: Pick the Right Kubernetes Abstraction with GitOps](https://itnext.io/stop-building-platforms-nobody-uses-pick-the-right-kubernetes-abstraction-with-gitops-64681357690f) - [From CI to Kubernetes Catalog: Building a Composable Platform with GitOps and vCluster](https://itnext.io/from-ci-to-kubernetes-catalog-building-a-composable-platform-with-gitops-and-vcluster-7e1decaa81da) - [Exploring Cloud Native projects in CNCF Sandbox. Part 3: 14 arrivals of 2024 H1](https://blog.palark.com/cncf-sandbox-2024-h1/) - [Descomplicando o Score (CNCF)](https://www.linkedin.com/pulse/descomplicando-o-score-cncf-clecio-antao-nyt3f/) - [Who’s Really Responsible for an Internal Developer Platform?](https://itnext.io/whos-really-responsible-for-an-internal-developer-platform-5dce5f2a0401) - [From Docker Compose to Score: A Platform Engineering Guide](https://waynegoosen.com/post/platform-engineering-transition-docker-compose-to-score-specification/) - [What if the `Azure-Samples/aks-store-demo` was using Score?](https://itnext.io/what-if-the-azure-samples-aks-store-demo-was-using-score-655c55f1c3dd) - [Score 容器化服务编排的抽象层](https://george.betterde.com/cn/devops/20240801/) - [Score: New CNCF Sandbox Tool for Infrastructure-Centric Dev](https://thenewstack.io/score-new-cncf-sandbox-tool-for-infrastructure-centric-dev/) - [Dapr with Score and Humanitec — Improving the Developer Experience of your Platform, on steroids!](https://medium.com/@mabenoit/dapr-with-score-and-humanitec-developer-experience-with-your-platform-on-steroids-a848f2de0a5a) - [Mastering Workload Specifications: Finding the Right Fit for Your Developers](https://www.infracloud.io/blogs/mastering-workload-specifications/) ## Leverage example projects using Score - [`mathieu-benoit/score-kro-demo`](https://github.com/mathieu-benoit/score-kro-demo) - [`mathieu-benoit/deploy-backstage-with-score`](https://github.com/mathieu-benoit/deploy-backstage-with-score) - [`la-cc/score-k8s-webapp`](https://github.com/la-cc/score-k8s-webapp) - [`mathieu-benoit/nginx-score-demo`](https://github.com/mathieu-benoit/nginx-score-demo) - [`kendallroden/kubecon-dapr-score-demo`](https://github.com/kendallroden/kubecon-dapr-score-demo) - [`score-spec/sample-score-app`](https://github.com/score-spec/sample-score-app) - [`score-spec/sample-app-gif`](https://github.com/score-spec/sample-app-gif) - [`Humanitec-DemoOrg/onlineboutique-demo`](https://github.com/Humanitec-DemoOrg/onlineboutique-demo) - [`Humanitec-DemoOrg/aks-store-demo`](https://github.com/Humanitec-DemoOrg/aks-store-demo) - [`mathieu-benoit/dapr-score`](https://github.com/mathieu-benoit/dapr-score-humanitec) --- ## File: content/en/docs/examples/included/backstage-score-compose.md --- title: "Backstage" headless: true toc_hide: true --- To begin, follow the [installation instructions](/docs/score-implementation/score-compose/installation) to install the latest version of `score-compose`. ## `init` Initialize your current `score-compose` workspace, run the following command in your terminal: ```bash score-compose init --no-sample \ --provisioners https://raw.githubusercontent.com/score-spec/community-provisioners/refs/heads/main/dns/score-compose/10-dns-with-url.provisioners.yaml ``` The `init` command will create the `.score-compose` directory with the [default resource provisioners]({{< relref "/docs/score-implementation/score-compose/resources-provisioners/" >}}) available. We are also importing one external file to support the `dns` dependencies: [`dns` provisioner](https://docs.score.dev/examples/resource-provisioners/community/dns/score-compose/template/dns-with-url/). You can see the resource provisioners available by running this command: ```bash score-compose provisioners list ``` The Score file example illustrated uses three resource types: `postgres-instance`, `dns` and `route`. ```none +-------------------+-------+------------------+--------------------------------+---------------------------------+ | TYPE | CLASS | PARAMS | OUTPUTS | DESCRIPTION | +-------------------+-------+------------------+--------------------------------+---------------------------------+ | dns | (any) | | host, url | Outputs a *.localhost domain as | | | | | | the hostname and associated URL | | | | | | in http on port 8080 | +-------------------+-------+------------------+--------------------------------+---------------------------------+ | postgres-instance | (any) | | host, password, port, username | Provisions a dedicated | | | | | | PostgreSQL instance | +-------------------+-------+------------------+--------------------------------+---------------------------------+ | route | (any) | host, path, port | | Provisions an Ingress route on | | | | | | a shared Nginx instance | +-------------------+-------+------------------+--------------------------------+---------------------------------+ ``` ## `generate` Convert the `score.yaml` file into a deployable `compose.yaml`, run the following command in your terminal: ```bash score-compose generate score.yaml --image ghcr.io/mathieu-benoit/backstage:latest ``` The `generate` command will add the input `score.yaml` workload with a particular container image to the `.score-compose/state.yaml` state file and generate the output `compose.yaml`. If you want to build the container image when this `compose.yaml` will be deployed, you can run this `generate` command with the `--build` parameter instead: ```bash score-compose generate score.yaml --build 'main={"context":".","tags":["backstage:local"]}' ``` See the generated `compose.yaml` by running this command: ```bash cat compose.yaml ``` If you make any modifications to the `score.yaml` file, run `score-compose generate score.yaml` to regenerate the output `compose.yaml`. ## `resources` Get the information of the resources dependencies of the workload, run the following command: ```bash score-compose resources list ``` ```none +----------------------------------------+--------------------------------+ | UID | OUTPUTS | +----------------------------------------+--------------------------------+ | dns.default#backstage.dns | host, url | +----------------------------------------+--------------------------------+ | postgres-instance.default#backstage.pg | host, password, port, username | +----------------------------------------+--------------------------------+ | route.default#backstage.route | | +----------------------------------------+--------------------------------+ ``` At this stage, we can already see the value of the `dns` resource generated: ```bash score-compose resources get-outputs dns.default#backstage.dns ``` ```none { "host": "dnsjdtv57.localhost", "url": "http://dnsjdtv57.localhost:8080" } ``` Same for the `postgres-instance` resource: ```bash score-compose resources get-outputs postgres-instance.default#backstage.pg ``` ```none { "host": "pg-OuoTNo", "password": "REDACTED", "port": 5432, "username": "REDACTED" } ``` ## `docker compose` Run `docker compose up` to execute the generated `compose.yaml` file: ```bash docker compose up -d ``` ```none [+] Running 4/4 ✔ Container deploy-backstage-with-score-pg-OuoTNo-1 Healthy ✔ Container deploy-backstage-with-score-wait-for-resources-1 Exited ✔ Container deploy-backstage-with-score-routing-6uaxax-1 Started ✔ Container deploy-backstage-with-score-backstage-backstage-1 Started ``` ## `docker ps` See the running containers: ```bash docker ps ``` ```none CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 9b2ff62333ff ghcr.io/mathieu-benoit/backstage:latest "node packages/backe…" 5 hours ago Up 5 hours deploy-backstage-with-score-backstage-backstage-1 4b1203acbf41 mirror.gcr.io/postgres:17-alpine "docker-entrypoint.s…" 5 hours ago Up 5 hours (healthy) 5432/tcp deploy-backstage-with-score-pg-mwgmNx-1 716b626ad841 mirror.gcr.io/nginx:1-alpine "/docker-entrypoint.…" 5 hours ago Up 5 hours 0.0.0.0:8080->80/tcp, [::]:8080->80/tcp deploy-backstage-with-score-routing-idouo7-1 ``` ## `curl localhost:8080` Test the running container, run the following command: ```bash curl localhost:8080 -H "Host: dnsjdtv57.localhost" ``` ```none ... Scaffolded Backstage App ... ``` Congrats! You’ve successfully deploy, with the `score-compose` implementation, a sample containerized Backstage workload talking to PostgreSQL and exposed via a DNS. You provisioned them through Docker, without writing the Docker Compose file by yourself. --- ## File: content/en/docs/examples/included/backstage-score-k8s.md --- title: "Backstage" headless: true toc_hide: true --- To begin, follow the [installation instructions](/docs/score-implementation/score-k8s/installation) to install the latest version of `score-k8s`. ## `init` Initialize your current `score-k8s` workspace, run the following command in your terminal: ```bash score-k8s init --no-sample \ --provisioners https://raw.githubusercontent.com/score-spec/community-provisioners/refs/heads/main/dns/score-k8s/10-dns-with-url.provisioners.yaml ``` The `init` command will create the `.score-k8s` directory with the [default resource provisioners]({{< relref "/docs/score-implementation/score-k8s/resources-provisioners/" >}}) available. We are also importing one external file to support the `dns` dependencies: [`dns` provisioner](https://docs.score.dev/examples/resource-provisioners/community/dns/score-k8s/template/dns-with-url/). You can see the resource provisioners available by running this command: ```bash score-k8s provisioners list ``` The Score file example illustrated uses three resource types: `postgres-instance`, `dns` and `route`. ```none +-------------------+-------+------------------+--------------------------------+---------------------------------+ | TYPE | CLASS | PARAMS | OUTPUTS | DESCRIPTION | +-------------------+-------+------------------+--------------------------------+---------------------------------+ | dns | (any) | | host, url | Outputs a *.localhost domain as | | | | | | the hostname and associated URL | | | | | | in http on port 80 | +-------------------+-------+------------------+--------------------------------+---------------------------------+ | postgres-instance | (any) | | host, password, port, username | Provisions a dedicated | | | | | | PostgreSQL instance | +-------------------+-------+------------------+--------------------------------+---------------------------------+ | route | (any) | host, path, port | | Provisions an Ingress route on | | | | | | a shared Nginx instance | +-------------------+-------+------------------+--------------------------------+---------------------------------+ ``` ## `generate` Convert the `score.yaml` file into a deployable `manifests.yaml`, run the following command in your terminal: ```bash score-k8s generate score.yaml --image ghcr.io/mathieu-benoit/backstage:latest ``` The `generate` command will add the input `score.yaml` workload with a particular container image to the `.score-k8s/state.yaml` state file and generate the output `manifests.yaml`. See the generated `manifests.yaml` by running this command: ```bash cat manifests.yaml ``` If you make any modifications to the `score.yaml` file, run `score-k8s generate score.yaml` to regenerate the output `manifests.yaml`. ## `resources` Get the information of the resources dependencies of the workload, run the following command: ```bash score-k8s resources list ``` ```none +----------------------------------------+--------------------------------+ | UID | OUTPUTS | +----------------------------------------+--------------------------------+ | dns.default#backstage.dns | host, url | +----------------------------------------+--------------------------------+ | postgres-instance.default#backstage.pg | host, password, port, username | +----------------------------------------+--------------------------------+ | route.default#backstage.route | | +----------------------------------------+--------------------------------+ ``` At this stage, we can already see the value of the `dns` resource generated: ```bash score-k8s resources get-outputs dns.default#backstage.dns ``` ```none { "host": "dnsnocrke.localhost", "url": "http://dnsnocrke.localhost:80" } ``` Same for the `postgres-instance` resource: ```bash score-k8s resources get-outputs postgres-instance.default#backstage.pg ``` ```none { "host": "pg-backstage-d7058793", "password": "REDACTED", "port": 5432, "username": "REDACTED" } ``` ## `kubectl apply` _Here you will need to have access to a Kubernetes cluster to execute the following commands. You can follow [these instructions](/docs/how-to/score-k8s/kind-cluster/) if you want to set up a Kind cluster._ Run `kubectl apply` to execute the generated `manifests.yaml` file: ```bash kubectl apply -f manifests.yaml ``` ```none secret/pg-backstage-d7058793 created statefulset.apps/pg-backstage-d7058793 created service/pg-backstage-d7058793 created httproute.gateway.networking.k8s.io/route-backstage-76d19d47 created service/backstage created deployment.apps/backstage created ``` ## `kubectl get all` See the running containers: ```bash kubectl get all,httproute ``` ```none NAME READY STATUS RESTARTS AGE pod/backstage-7667f68bf9-vnlw9 1/1 Running 0 37s pod/pg-backstage-d7058793-0 1/1 Running 0 37s NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE service/backstage ClusterIP 10.96.146.148 7007/TCP 37s service/pg-backstage-d7058793 ClusterIP 10.96.21.104 5432/TCP 37s NAME READY UP-TO-DATE AVAILABLE AGE deployment.apps/backstage 1/1 1 1 37s NAME DESIRED CURRENT READY AGE replicaset.apps/backstage-7667f68bf9 1 1 1 37s NAME READY AGE statefulset.apps/pg-backstage-d7058793 1/1 37s NAME HOSTNAMES AGE httproute.gateway.networking.k8s.io/route-backstage-76d19d47 ["dnsnocrke.localhost"] 37s ``` ## `curl localhost` Test the running container, run the following command: ```bash curl localhost -H "Host: dnsnocrke.localhost" ``` ```none ... Scaffolded Backstage App ... ``` Congrats! You’ve successfully deploy, with the `score-k8s` implementation, a sample containerized Backstage workload talking to PostgreSQL and exposed via a DNS. You provisioned them through `kubectl`, without writing the Kubernetes manifests file by yourself. --- ## File: content/en/docs/examples/included/dapr-score-compose.md --- title: "Dapr" headless: true toc_hide: true --- To begin, follow the [installation instructions](/docs/score-implementation/score-compose/installation) to install the latest version of `score-compose`. ## `init` Initialize your current `score-compose` workspace, run the following command in your terminal: ```bash score-compose init --no-sample \ --provisioners https://raw.githubusercontent.com/score-spec/community-provisioners/refs/heads/main/dapr-state-store/score-compose/10-redis-dapr-state-store.provisioners.yaml \ --patch-templates https://raw.githubusercontent.com/score-spec/community-patchers/refs/heads/main/score-compose/dapr.tpl ``` The `init` command will create the `.score-compose` directory with the [default resource provisioners]({{< relref "/docs/score-implementation/score-compose/resources-provisioners/" >}}) available. We are also importing one external file to seamlessly generate a Dapr StateStore `Component` pointing to a Redis database: [`dapr-state-store` provisioner](https://docs.score.dev/examples/resource-provisioners/community/dapr-state-store/score-compose/template/redis-dapr-state-store/). You can see the resource provisioners available by running this command: ```bash score-compose provisioners list ``` The Score file example illustrated uses three resource types: `dapr-state-store`, `dns` and `route`. ```none +-------------------+-------+------------------+--------------------------------+---------------------------------+ | TYPE | CLASS | PARAMS | OUTPUTS | DESCRIPTION | +-------------------+-------+------------------+--------------------------------+---------------------------------+ | dapr-state-store | (any) | | name | Generates a Dapr StateStore | | | | | | Component pointing to a Redis | | | | | | Service. | +-------------------+-------+------------------+--------------------------------+---------------------------------+ | dns | (any) | | host | Outputs a *.localhost domain | | | | | | as the hostname | +-------------------+-------+------------------+--------------------------------+---------------------------------+ | route | (any) | host, path, port | | Provisions an Ingress route on | | | | | | a shared Nginx instance | +-------------------+-------+------------------+--------------------------------+---------------------------------+ ``` By using the [`--patch-templates`](/docs/score-implementation/score-compose/patch-templates/) (in this case: [`dapr.tpl`](https://github.com/score-spec/community-patchers/blob/main/score-compose/dapr.tpl)) we are seamlessly generating the Dapr `scheduler` and `placement` containers in addition to a Dapr `Sidecar` container for any Workload. ## `generate` Convert the `score.yaml` file into a deployable `compose.yaml`, run the following command in your terminal: ```bash score-compose generate score.yaml --image ghcr.io/dapr/samples/hello-k8s-node:latest ``` The `generate` command will add the input `score.yaml` workload with a particular container image to the `.score-compose/state.yaml` state file and generate the output `compose.yaml`. If you want to build the container image when this `compose.yaml` will be deployed, you can run this `generate` command with the `--build` parameter instead: ```bash score-compose generate score.yaml --build 'main={"context":".","tags":["hello-dapr-node:local"]}' ``` See the generated `compose.yaml` by running this command: ```bash cat compose.yaml ``` If you make any modifications to the `score.yaml` file, run `score-compose generate score.yaml` to regenerate the output `compose.yaml`. ## `resources` Get the information of the resources dependencies of the workload, run the following command: ```bash score-compose resources list ``` ```none +----------------------------------------------+---------+ | UID | OUTPUTS | +----------------------------------------------+---------+ | dapr-state-store.default#nodeapp.state-store | name | +----------------------------------------------+---------+ | dns.default#nodeapp.dns | host | +----------------------------------------------+---------+ | route.default#nodeapp.route | | +----------------------------------------------+---------+ ``` At this stage, we can already see the value of the `dns` resource generated: ```bash score-compose resources get-outputs dns.default#nodeapp.dns --format '{{ .host }}' ``` ```none dnsbcsqnd.localhost ``` Same for the `dapr-state-store` resource: ```bash score-compose resources get-outputs dapr-state-store.default#nodeapp.state-store ``` ```none { "name": "redis-PANgkO" } ``` ## `docker compose` Run `docker compose up` to execute the generated `compose.yaml` file: ```bash docker compose up -d ``` ```none [+] Running 7/7 ✔ Container dapr-redis-PANgkO-1 Started ✔ Container dapr-placement-1 Started ✔ Container dapr-scheduler-1 Started ✔ Container dapr-routing-aPCJB9-1 Started ✔ Container dapr-wait-for-resources-1 Exited ✔ Container dapr-nodeapp-nodeapp-1 Started ✔ Container dapr-nodeapp-nodeapp-sidecar-1 Started ``` ## `docker ps` See the running containers: ```bash docker ps ``` ```none CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 3c7dfd658b1a mirror.gcr.io/redis:7-alpine "redis-server /usr/l…" About a minute ago Up About a minute 6379/tcp dapr-redis-PANgkO-1 85c836ae19d2 ghcr.io/dapr/daprd:latest "./daprd --app-id=no…" 8 hours ago Up 59 seconds dapr-nodeapp-nodeapp-sidecar-1 1f7affe5d910 ghcr.io/dapr/samples/hello-k8s-node:latest "docker-entrypoint.s…" 8 hours ago Up About a minute 3000/tcp dapr-nodeapp-nodeapp-1 e7e1ae181a28 mirror.gcr.io/nginx:1-alpine "/docker-entrypoint.…" 8 hours ago Up About a minute 0.0.0.0:8080->80/tcp, [::]:8080->80/tcp dapr-routing-aPCJB9-1 f2983452962a ghcr.io/dapr/scheduler:latest "./scheduler --port …" 8 hours ago Up About a minute 0.0.0.0:50007->50007/tcp, :::50007->50007/tcp dapr-scheduler-1 913af9a5c8fc ghcr.io/dapr/placement:latest "./placement --port …" 8 hours ago Up About a minute 0.0.0.0:50006->50006/tcp, :::50006->50006/tcp dapr-placement-1 ``` ## `curl localhost:8080` Test the running container, run the following command: ```bash curl localhost:8080 -H "Host: dnsbcsqnd.localhost" ``` This will get the expected error showing that the container is successfully running: ```none Cannot GET / ``` You can check the logs of the running container: ```none Node App listening on port 3000! ``` Congrats! You’ve successfully deploy, with the `score-compose` implementation, a sample containerized NodeJS workload talking to a Dapr StateStore (Redis) and exposed via a DNS. You provisioned them through Docker, without writing the Docker Compose file by yourself. --- ## File: content/en/docs/examples/included/dapr-score-k8s.md --- title: "Dapr" headless: true toc_hide: true --- To begin, follow the [installation instructions](/docs/score-implementation/score-k8s/installation) to install the latest version of `score-k8s`. ## `init` Initialize your current `score-k8s` workspace, run the following command in your terminal: ```bash score-k8s init --no-sample \ --provisioners https://raw.githubusercontent.com/score-spec/community-provisioners/refs/heads/main/dapr-state-store/score-k8s/10-redis-dapr-state-store.provisioners.yaml ``` The `init` command will create the `.score-k8s` directory with the [default resource provisioners]({{< relref "/docs/score-implementation/score-k8s/resources-provisioners/" >}}) available. We are also importing one external file to generate a Dapr StateStore `Component` pointing to a Redis database: [`dapr-state-store` provisioner](https://docs.score.dev/examples/resource-provisioners/community/dapr-state-store/score-k8s/template/redis/). You can see the resource provisioners available by running this command: ```bash score-k8s provisioners list ``` The Score file example illustrated uses three resource types: `dapr-state-store`, `dns` and `route`. ```none +------------------+-------+------------------+--------------------------------+---------------------------------+ | TYPE | CLASS | PARAMS | OUTPUTS | DESCRIPTION | +------------------+-------+------------------+--------------------------------+---------------------------------+ | dapr-state-store | (any) | | name | Generates a Dapr StateStore | | | | | | Component pointing to a Redis | | | | | | StatefulSet. | +------------------+-------+------------------+--------------------------------+---------------------------------+ | dns | (any) | | host | Outputs a *.localhost domain | | | | | | as the hostname | +------------------+-------+------------------+--------------------------------+---------------------------------+ | route | (any) | host, path, port | | Provisions an Ingress route on | | | | | | a shared Nginx instance | +------------------+-------+------------------+--------------------------------+---------------------------------+ ``` ## `generate` Convert the `score.yaml` file into a deployable `manifests.yaml`, run the following command in your terminal: ```bash score-k8s generate score.yaml --image ghcr.io/dapr/samples/hello-k8s-node:latest ``` The `generate` command will add the input `score.yaml` workload with a particular container image to the `.score-k8s/state.yaml` state file and generate the output `manifests.yaml`. See the generated `manifests.yaml` by running this command: ```bash cat manifests.yaml ``` If you make any modifications to the `score.yaml` file, run `score-k8s generate score.yaml` to regenerate the output `manifests.yaml`. ## `resources` Get the information of the resources dependencies of the workload, run the following command: ```bash score-k8s resources list ``` ```none +----------------------------------------------+---------+ | UID | OUTPUTS | +----------------------------------------------+---------+ | dapr-state-store.default#nodeapp.state-store | name | +----------------------------------------------+---------+ | dns.default#nodeapp.dns | host | +----------------------------------------------+---------+ | route.default#nodeapp.route | | +----------------------------------------------+---------+ ``` At this stage, we can already see the value of the `dns` resource generated: ```bash score-k8s resources get-outputs dns.default#nodeapp.dns --format '{{ .host }}' ``` ```none dnsutripw.localhost ``` Same for the `dapr-state-store` resource: ```bash score-k8s resources get-outputs dapr-state-store.default#nodeapp.state-store ``` ```none { "name": "redis-nodeapp-58f89990" } ``` ## `kubectl apply` _Here you will need to have access to a Kubernetes cluster to execute the following commands. You can follow [these instructions](/docs/how-to/score-k8s/kind-cluster/) if you want to set up a Kind cluster. Your Kubernetes cluster should also have [Dapr installed](https://docs.dapr.io/operations/hosting/kubernetes/kubernetes-overview/) in it._ Run `kubectl apply` to execute the generated `manifests.yaml` file: ```bash kubectl apply -f manifests.yaml ``` ```none secret/redis-nodeapp-58f89990 created statefulset.apps/redis-nodeapp-58f89990 created service/redis-nodeapp-58f89990 created component.dapr.io/redis-nodeapp-58f89990 created httproute.gateway.networking.k8s.io/route-nodeapp-3f3a9362 created service/nodeapp created deployment.apps/nodeapp created ``` ## `kubectl get all` See the running containers: ```bash kubectl get all ``` ```none NAME READY STATUS RESTARTS AGE pod/nodeapp-859d5458f6-75wjj 1/1 Running 0 3m43s pod/redis-nodeapp-58f89990-0 1/1 Running 0 3m43s NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE service/nodeapp ClusterIP 10.96.138.145 3000/TCP 3m43s service/nodeapp-dapr ClusterIP None 80/TCP,50001/TCP,50002/TCP,9090/TCP 51s service/redis-nodeapp-58f89990 ClusterIP 10.96.71.72 6379/TCP 3m43s NAME READY UP-TO-DATE AVAILABLE AGE deployment.apps/nodeapp 1/1 1 1 3m43s NAME DESIRED CURRENT READY AGE replicaset.apps/nodeapp-859d5458f6 1 1 1 3m43s NAME READY AGE statefulset.apps/redis-nodeapp-58f89990 1/1 3m43s ``` ## `curl localhost` Test the running container, run the following command: ```bash curl localhost -H "Host: dnsutripw.localhost" ``` This will get the expected error showing that the container is successfully running: ```none Cannot GET / ``` You can check the logs of the running container: ```none Node App listening on port 3000! ``` Congrats! You’ve successfully deploy, with the `score-k8s` implementation, a sample containerized NodeJS workload talking to a Dapr StateStore (Redis) and exposed via a DNS. You provisioned them through `kubectl`, without writing the Kubernetes manifests file by yourself. --- ## File: content/en/docs/examples/included/frontend-and-backend-score-compose.md --- title: "Frontend & Backend" headless: true toc_hide: true --- To begin, follow the [installation instructions](/docs/score-implementation/score-compose/installation) to install the latest version of `score-compose`. ## `init` Initialize your current `score-compose` workspace, run the following command in your terminal: ```bash score-compose init --no-sample \ --provisioners https://raw.githubusercontent.com/score-spec/community-provisioners/refs/heads/main/service/score-compose/10-service.provisioners.yaml \ --provisioners https://raw.githubusercontent.com/score-spec/community-provisioners/refs/heads/main/dns/score-compose/10-dns-with-url.provisioners.yaml ``` The `init` command will create the `.score-compose` directory with the [default resource provisioners]({{< relref "/docs/score-implementation/score-compose/resources-provisioners/" >}}) available. We are also importing two external files to support the `dns` dependencies: [`dns` provisioner](https://docs.score.dev/examples/resource-provisioners/community/dns/score-compose/template/dns-with-route/) and `service` dependencies: [`service` provisioner](https://docs.score.dev/examples/resource-provisioners/community/service/score-compose/template/static-service/). You can see the resource provisioners available by running this command: ```bash score-compose provisioners list ``` The Score file example illustrated uses three resource types: `postgres-instance`, `dns` and `route`. ``` /* Detailed source-code truncated for AI context efficiency. */ ``` ## `generate` Convert the `score-frontend.yaml` and `score-backend.yaml` files into a deployable `compose.yaml`, run the following command in your terminal: ```bash score-compose generate score-backend.yaml \ --image ghcr.io/mathieu-benoit/backstage-backend:latest score-compose generate score-frontend.yaml \ --image ghcr.io/mathieu-benoit/backstage-frontend:latest ``` The `generate` command will add the input `score-frontend.yaml` and `score-backend.yaml` workloads with a particular container image to the `.score-compose/state.yaml` state file and generate the output `compose.yaml`. If you want to build the container image when this `compose.yaml` will be deployed, you can run this `generate` command with the `--build` parameter instead: ```bash score-compose generate score-backend.yaml --build 'main={"context":".","tags":["backend:local"]}' ``` See the generated `compose.yaml` by running this command: ```bash cat compose.yaml ``` If you make any modifications to the `score-frontend.yaml` or `score-backend.yaml` files, run `score-compose generate score.yaml` to regenerate the output `compose.yaml`. ## `resources` Get the information of the resources dependencies of the workload, run the following command: ```bash score-compose resources list ``` ```none +--------------------------------------+--------------------------------+ | UID | OUTPUTS | +--------------------------------------+--------------------------------+ | dns.default#dns | host, url | +--------------------------------------+--------------------------------+ | postgres-instance.default#backend.pg | host, password, port, username | +--------------------------------------+--------------------------------+ | service.default#frontend.backend | name | +--------------------------------------+--------------------------------+ | route.default#backend.route | | +--------------------------------------+--------------------------------+ | route.default#frontend.route | | +--------------------------------------+--------------------------------+ ``` At this stage, we can already see the value of the `dns` resource generated: ```bash score-compose resources get-outputs dns.default#dns ``` ```none { "host": "dnsjdtv57.localhost", "url": "http://dnsjdtv57.localhost:8080" } ``` Same for the `postgres-instance` resource: ```bash score-compose resources get-outputs postgres-instance.default#backend.pg ``` ```none { "host": "pg-OuoTNo", "password": "REDACTED", "port": 5432, "username": "REDACTED" } ``` ## `docker compose` Run `docker compose up` to execute the generated `compose.yaml` file: ```bash docker compose up -d ``` ```none [+] Running 5/5 ✔ Container deploy-backstage-with-score-routing-5QD6z8-1 Running ✔ Container deploy-backstage-with-score-pg-glZvLw-1 Healthy ✔ Container deploy-backstage-with-score-wait-for-resources-1 Exited ✔ Container deploy-backstage-with-score-backend-backend-1 Running ✔ Container deploy-backstage-with-score-frontend-frontend-1 Running ``` ## `docker ps` See the running containers: ```bash docker ps ``` ```none CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 1712a2002838 ghcr.io/mathieu-benoit/backstage-frontend:latest "/docker-entrypoint.…" 3 minutes ago Up 2 minutes 80/tcp deploy-backstage-with-score-frontend-frontend-1 6bf734ab9179 ghcr.io/mathieu-benoit/backstage-backend:latest "node packages/backe…" 3 minutes ago Up 2 minutes -backend-backend-1 9a126ed0456c mirror.gcr.io/nginx:1-alpine "/docker-entrypoint.…" 3 minutes ago Up 2 minutes 0.0.0.0:8080->80/tcp, [::]:8080->80/tcp deploy-backstage-with-score-routing-5QD6z8-1 5ec1732b6695 mirror.gcr.io/postgres:17-alpine "docker-entrypoint.s…" 3 minutes ago Up 3 minutes (healthy) 5432/tcp -pg-glZvLw-1 ``` ## `curl localhost:8080` Test the running container, run the following command: ```bash curl localhost:8080 -H "Host: dnsjdtv57.localhost" ``` ```none ... Scaffolded Backstage App ... ``` Congrats! You’ve successfully deploy, with the `score-compose` implementation, a containerized Frontend application talking to a containerized Backend application exposed via a DNS. You provisioned them through Docker, without writing the Docker Compose file by yourself. --- ## File: content/en/docs/examples/included/frontend-and-backend-score-k8s.md --- title: "Frontend & Backend" headless: true toc_hide: true --- To begin, follow the [installation instructions](/docs/score-implementation/score-k8s/installation) to install the latest version of `score-k8s`. ## `init` Initialize your current `score-k8s` workspace, run the following command in your terminal: ```bash score-k8s init --no-sample \ --provisioners https://raw.githubusercontent.com/score-spec/community-provisioners/refs/heads/main/service/score-k8s/10-service.provisioners.yaml \ --provisioners https://raw.githubusercontent.com/score-spec/community-provisioners/refs/heads/main/dns/score-k8s/10-dns-with-url.provisioners.yaml ``` The `init` command will create the `.score-k8s` directory with the [default resource provisioners]({{< relref "/docs/score-implementation/score-k8s/resources-provisioners/" >}}) available. We are also importing two external files to support the `dns` dependencies: [`dns` provisioner](https://docs.score.dev/examples/resource-provisioners/community/dns/score-k8s/template/dns-with-url/) and `service` dependencies: [`service` provisioner](https://docs.score.dev/examples/resource-provisioners/community/service/score-k8s/template/static-service/). You can see the resource provisioners available by running this command: ```bash score-k8s provisioners list ``` The Score file example illustrated uses three resource types: `postgres-instance`, `dns` and `route`. ``` /* Detailed source-code truncated for AI context efficiency. */ ``` ## `generate` Convert the `score-frontend.yaml` and `score-backend.yaml` files into a deployable `manifests.yaml`, run the following command in your terminal: ```bash score-k8s generate score-backend.yaml \ --image ghcr.io/mathieu-benoit/backstage-backend:latest score-k8s generate score-frontend.yaml \ --image ghcr.io/mathieu-benoit/backstage-frontend:latest ``` The `generate` command will add the input `score-frontend.yaml` and `score-backend.yaml` workloads with a particular container image to the `.score-k8s/state.yaml` state file and generate the output `manifests.yaml`. See the generated `manifests.yaml` by running this command: ```bash cat manifests.yaml ``` If you make any modifications to the `score-frontend.yaml` or `score-backend.yaml` files, run `score-k8s generate score.yaml` to regenerate the output `manifests.yaml`. ## `resources` Get the information of the resources dependencies of the workload, run the following command: ```bash score-k8s resources list ``` ```none +--------------------------------------+--------------------------------+ | UID | OUTPUTS | +--------------------------------------+--------------------------------+ | dns.default#dns | host, url | +--------------------------------------+--------------------------------+ | postgres-instance.default#backend.pg | host, password, port, username | +--------------------------------------+--------------------------------+ | service.default#frontend.backend | name | +--------------------------------------+--------------------------------+ | route.default#backend.route | | +--------------------------------------+--------------------------------+ | route.default#frontend.route | | +--------------------------------------+--------------------------------+ ``` At this stage, we can already see the value of the `dns` resource generated: ```bash score-k8s resources get-outputs dns.default#dns ``` ```none { "host": "dnsnocrke.localhost", "url": "http://dnsnocrke.localhost:80" } ``` Same for the `postgres-instance` resource: ```bash score-k8s resources get-outputs postgres-instance.default#backend.pg ``` ```none { "host": "pg-backstage-d7058793", "password": "REDACTED", "port": 5432, "username": "REDACTED" } ``` ## `kubectl apply` _Here you will need to have access to a Kubernetes cluster to execute the following commands. You can follow [these instructions](/docs/how-to/score-k8s/kind-cluster/) if you want to set up a Kind cluster._ Run `kubectl apply` to execute the generated `manifests.yaml` file: ```bash kubectl apply -f manifests.yaml ``` ```none secret/pg-backend-61f0e7c1 created statefulset.apps/pg-backend-61f0e7c1 created service/pg-backend-61f0e7c1 created httproute.gateway.networking.k8s.io/route-backend-711726ce created httproute.gateway.networking.k8s.io/route-frontend-61f9e1b8 created service/backend created deployment.apps/backend created service/frontend created deployment.apps/frontend created ``` ## `kubectl get all` See the running containers: ```bash kubectl get all,httproute ``` ```none NAME READY STATUS RESTARTS AGE pod/backend-57fc5664d-4hg7l 1/1 Running 0 22s pod/frontend-db644cf86-n9cjd 1/1 Running 0 22s pod/pg-backend-61f0e7c1-0 1/1 Running 0 60m NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE service/backend ClusterIP 10.96.183.51 7007/TCP 60m service/frontend ClusterIP 10.96.90.135 3000/TCP 60m service/pg-backend-61f0e7c1 ClusterIP 10.96.45.108 5432/TCP 60m NAME READY UP-TO-DATE AVAILABLE AGE deployment.apps/backend 1/1 1 1 60m deployment.apps/frontend 1/1 1 1 60m NAME DESIRED CURRENT READY AGE replicaset.apps/backend-57fc5664d 1 1 1 22s replicaset.apps/backend-64fc859b9d 0 0 0 60m replicaset.apps/frontend-68ccb68884 0 0 0 60m replicaset.apps/frontend-db644cf86 1 1 1 22s NAME READY AGE statefulset.apps/pg-backend-61f0e7c1 1/1 60m NAME HOSTNAMES AGE httproute.gateway.networking.k8s.io/route-backend-711726ce ["dnsxmfazk.localhost"] 60m httproute.gateway.networking.k8s.io/route-frontend-61f9e1b8 ["dnsxmfazk.localhost"] 60m ``` ## `curl localhost` Test the running container, run the following command: ```bash curl localhost -H "Host: dnsnocrke.localhost" ``` ```none ... Scaffolded Backstage App ... ``` Congrats! You’ve successfully deploy, with the `score-k8s` implementation, a containerized Frontend application talking to a containerized Backend application exposed via a DNS. You provisioned them through `kubectl`, without writing the Kubernetes manifests file by yourself. --- ## File: content/en/docs/examples/included/llm-models-dmr.md --- title: "LLM Models with Docker Model Runner" headless: true toc_hide: true --- To begin, follow the [installation instructions](/docs/score-implementation/score-compose/installation) to install the latest version of `score-compose`. {{% alert title="Prerequisites" %}} Docker Model Runner (DMR) needs to be set up in your local environment. Follow the [DMR get started guide](https://docs.docker.com/model-runner/get-started/) to set it up. {{% /alert %}} ## `init` Initialize your current `score-compose` workspace with the Docker Model Runner provisioner. Run the following command in your terminal: ```bash score-compose init --no-sample \ --provisioners https://raw.githubusercontent.com/score-spec/community-provisioners/refs/heads/main/llm-model/score-compose/10-dmr-llm-model.provisioners.yaml ``` The `init` command will create the `.score-compose` directory with the [default resource provisioners]({{< relref "/docs/score-implementation/score-compose/resources-provisioners/" >}}) available, plus the Docker Model Runner-specific provisioner for the `llm-model` resource type. You can see the resource provisioners available by running this command: ```bash score-compose provisioners list ``` The Score file example illustrated uses two resource types: `llm-model` and `volume`. ```none +---------------+-------+--------+--------------------+---------------------------------+ | TYPE | CLASS | PARAMS | OUTPUTS | DESCRIPTION | +---------------+-------+--------+--------------------+---------------------------------+ | llm-model | (any) | model | url | Provisions an LLM model via | | | | | | Docker Model Runner | +---------------+-------+--------+--------------------+---------------------------------+ | volume | (any) | | source, type | Creates a persistent volume | | | | | | that can be mounted on a | | | | | | workload. | +---------------+-------+--------+--------------------+---------------------------------+ ``` ## `generate` Convert the `score.yaml` file into a deployable `compose.yaml`. Run the following command in your terminal: ```bash score-compose generate score.yaml \ --image ghcr.io/open-webui/open-webui:main-slim \ --publish 8080:open-webui:8080 \ --override-property containers.open-webui.variables.WEBUI_NAME="Hello, DMR with Score Compose!" \ --output compose.yaml ``` The `generate` command will add the input `score.yaml` workload with a particular container image to the `.score-compose/state.yaml` state file and generate the output `compose.yaml`. The `--publish` flag exposes the Open WebUI port to the host so you can access it in your browser. See the generated `compose.yaml` by running this command: ```bash cat compose.yaml ``` If you make any modifications to the `score.yaml` file, run `score-compose generate score.yaml` to regenerate the output `compose.yaml`. ## `docker compose` Run `docker compose up` to execute the generated `compose.yaml` file: ```bash docker compose up -d --wait ``` This will pull the requested LLM models (`ai/gemma3:270M-UD-IQ2_XXS` and `ai/smollm2:135M-Q2_K`) via Docker Model Runner, and then start the Open WebUI container connected to the DMR backend. ## Verify Check the Docker images, including the LLM model images pulled by Docker Model Runner: ```bash docker images ``` ```none REPOSITORY TAG SIZE open-webui main-slim ... ai/gemma3 270M-UD-IQ2_XXS 182MB ai/smollm2 135M-Q2_K 56.9MB ``` See the running containers: ```bash docker ps ``` You should see the `open-webui` container running and connected to the Docker Model Runner backend. ## Access Open WebUI Open your browser and navigate to [http://localhost:8080](http://localhost:8080) to access the Open WebUI frontend. You can now chat with the LLM models you deployed! Congrats! You've successfully deployed, with the `score-compose` implementation, local LLM models with Open WebUI and Docker Model Runner. You provisioned them through Docker, without writing the Docker Compose file by yourself. --- ## File: content/en/docs/examples/included/llm-models-ollama.md --- title: "LLM Models with Ollama" headless: true toc_hide: true --- To begin, follow the [installation instructions](/docs/score-implementation/score-compose/installation) to install the latest version of `score-compose`. ## `init` Initialize your current `score-compose` workspace with the Ollama provisioner and patch template. Run the following command in your terminal: ```bash score-compose init --no-sample \ --patch-templates https://raw.githubusercontent.com/score-spec/community-patchers/refs/heads/main/score-compose/ollama.tpl \ --provisioners https://raw.githubusercontent.com/score-spec/community-provisioners/refs/heads/main/llm-model/score-compose/10-ollama-llm-model-service.provisioners.yaml ``` The `init` command will create the `.score-compose` directory with the [default resource provisioners]({{< relref "/docs/score-implementation/score-compose/resources-provisioners/" >}}) available, plus the Ollama-specific provisioner for the `llm-model` resource type. The `--patch-templates` option adds the [`ollama.tpl`](https://github.com/score-spec/community-patchers/blob/main/score-compose/ollama.tpl) patch template which configures the Ollama service integration in the generated Docker Compose file. You can see the resource provisioners available by running this command: ```bash score-compose provisioners list ``` The Score file example illustrated uses two resource types: `llm-model` and `volume`. ```none +---------------+-------+--------+--------------------+---------------------------------+ | TYPE | CLASS | PARAMS | OUTPUTS | DESCRIPTION | +---------------+-------+--------+--------------------+---------------------------------+ | llm-model | (any) | model | url | Provisions an LLM model via | | | | | | Ollama | +---------------+-------+--------+--------------------+---------------------------------+ | volume | (any) | | source, type | Creates a persistent volume | | | | | | that can be mounted on a | | | | | | workload. | +---------------+-------+--------+--------------------+---------------------------------+ ``` ## `generate` Convert the `score.yaml` file into a deployable `compose.yaml`. Run the following command in your terminal: ```bash score-compose generate score.yaml \ --image ghcr.io/open-webui/open-webui:main-slim \ --publish 8080:open-webui:8080 \ --override-property containers.open-webui.variables.WEBUI_NAME="Hello, Ollama with Score Compose!" \ --output compose.yaml ``` The `generate` command will add the input `score.yaml` workload with a particular container image to the `.score-compose/state.yaml` state file and generate the output `compose.yaml`. The `--publish` flag exposes the Open WebUI port to the host so you can access it in your browser. See the generated `compose.yaml` by running this command: ```bash cat compose.yaml ``` If you make any modifications to the `score.yaml` file, run `score-compose generate score.yaml` to regenerate the output `compose.yaml`. ## `docker compose` Run `docker compose up` to execute the generated `compose.yaml` file: ```bash docker compose up -d --wait ``` This will pull the Ollama container image, start the Ollama service, pull the requested LLM models (`gemma3:270m` and `smollm2:135m`), and then start the Open WebUI container connected to the Ollama backend. ## Verify Check the Docker images that were pulled: ```bash docker images ``` ```none REPOSITORY TAG SIZE ollama/ollama latest 3.33GB open-webui main-slim ... ``` Check that the LLM models were successfully pulled by Ollama: ```bash curl localhost:11434/api/tags | jq -r .models[].name ``` ```none gemma3:270m smollm2:135m ``` See the running containers: ```bash docker ps ``` You should see the `open-webui` and `ollama` containers running, along with some exited "puller" containers that were used to download the models. ## Access Open WebUI Open your browser and navigate to [http://localhost:8080](http://localhost:8080) to access the Open WebUI frontend. You can now chat with the LLM models you deployed! Congrats! You've successfully deployed, with the `score-compose` implementation, local LLM models with Open WebUI and Ollama. You provisioned them through Docker, without writing the Docker Compose file by yourself.