[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7ED4CDMFMG2MBV9JZ2KX2B6":3,"skill-same:01M7ED4CDMFMG2MBV9JZ2KX2B6":138},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":38,"commitSha":39,"skillDirectory":40,"forksCount":41,"repositoryHtmlUrl":42,"repositoryHomepageUrl":43,"ownerHtmlUrl":44,"ownerBlogUrl":45,"ownerDescription":46,"updatedAt":47,"install":48,"resources":61},"01M7ED4CDMFMG2MBV9JZ2KX2B6","aiq-deploy",null,"Use when asked to install, deploy, run, validate, troubleshoot, or stop NVIDIA AI-Q Blueprint infrastructure.","kael-odin\u002Fcodex-plugins-zh","kael-odin","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F166499071?v=4","01M7EBHJT87XN88WT3PEQ8ZGZK","Kael Odin","01M7EBHJVC91GD7PGJN3ZB14SG",37,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: aiq-deploy\ndescription: |\n  Use when asked to install, deploy, run, validate, troubleshoot, or stop NVIDIA AI-Q Blueprint infrastructure.\nlicense: Apache-2.0\ncompatibility: |\n  Designed for Claude Code, OpenCode, Codex, and Agent Skills-compatible tools. Requires Git, network\n  access to GitHub, and one selected runtime path: Docker Compose v2 for the default local deployment,\n  Python 3.11+ and uv for local process or CLI mode, Node.js 20+ and npm for local web UI mode, or\n  kubectl 1.28+ and Helm 3.12+ for Kubernetes and Helm mode.\nmetadata:\n  version: \"2.1.0\"\n  author: \"NVIDIA AI-Q Blueprint Team \u003Caiq-blueprint@nvidia.com>\"\n  github-url: \"https:\u002F\u002Fgithub.com\u002FNVIDIA-AI-Blueprints\u002Faiq\"\n  tags:\n    - nvidia\n    - aiq\n    - blueprint\n    - deploy\n    - operations\n    - agent-skills\nallowed-tools: Read Bash\n---\n\n# AIQ Deploy Skill\n\n## Purpose\n\nUse this skill to get a local or self-hosted NVIDIA AI-Q Blueprint server running and verified for use by\n`aiq-research`.\n\nThis skill owns setup, deployment, operational checks, troubleshooting, and shutdown. It does not run deep\nresearch itself. After deployment is healthy, hand off the verified server URL to `aiq-research`.\nThe workflow stays explicit so deployment validation and handoff are repeatable across supported agent clients.\n\n## Prerequisites\n\nUsers need:\n\n- Access to clone or update `https:\u002F\u002Fgithub.com\u002FNVIDIA-AI-Blueprints\u002Faiq`.\n- Git available in the shell.\n- One deployment runtime:\n  - Docker Engine with Docker Compose v2 for the default durable local deployment.\n  - Python 3.11+ and `uv` for local process or CLI mode.\n  - Node.js 20+ and `npm` for local browser UI development mode.\n  - `kubectl` 1.28+, Helm 3.12+, and access to a Kubernetes cluster for Helm mode.\n- Network access to GitHub, NVIDIA-hosted model endpoints, and any selected search provider.\n- Credentials stored outside chat. Hosted-model usage requires `NVIDIA_API_KEY`; web research requires at least\n  one supported search provider key such as `TAVILY_API_KEY`, `SERPER_API_KEY`, or `EXA_API_KEY`.\n- System capacity for the selected runtime. Docker Compose mode starts the AI-Q backend and PostgreSQL by default;\n  browser UI mode also uses frontend port `3000`. Self-hosted model or RAG deployments may require GPU resources.\n\nBefore writing secrets, verify `deploy\u002F.env` is ignored:\n\n```bash\ngit check-ignore deploy\u002F.env\n```\n\nExpected output: `deploy\u002F.env` or a matching ignore rule. If it is not ignored, stop and fix the ignore rule before\nplacing credentials in the file.\n\n## Instructions\n\n1. Locate or clone the AI-Q repository.\n2. Confirm the expected repository files exist.\n3. Select the deployment mode.\n4. Prepare `deploy\u002F.env` without overwriting user secrets.\n5. Check runtime prerequisites for the selected path.\n6. Start the selected deployment.\n7. Run basic validation.\n8. Report the verified `AIQ_SERVER_URL` for `aiq-research`.\n9. Ask whether to run optional deep research completion validation.\n\n### Step 1 - Locate or clone AI-Q\n\nIf no AI-Q checkout exists, read `references\u002Flocate-or-clone.md` before cloning. In an existing checkout, confirm the\nrequired files:\n\n```bash\npwd\ntest -f pyproject.toml\ntest -f deploy\u002F.env.example\ntest -d configs\n```\n\nExpected output: `pwd` prints the AI-Q repository path; the `test` commands exit with status 0 and no output.\n\n### Step 2 - Select the deployment mode\n\nIf the user asks to install, deploy, set up, or run AI-Q without naming a mode, ask:\n\n```text\nHow do you want to run AI-Q?\n\n1. Skill backend - backend-only service for aiq-research w\u002Fo browser UI.\n2. CLI - interactive terminal AI-Q.\n3. UI - browser AI-Q app with backend and frontend.\n4. Custom - choose an existing AI-Q config or review advanced customization docs before deployment.\n```\n\nWait for the user's answer before starting services.\n\nDo not ask this question when the user already specified a mode, such as Docker Compose, Helm, UI, CLI, or Agent Skill\nbackend. Do not ask the full mode question when `aiq-research` routed here because a deep research request needs a\nbackend. In that case, prefer Agent Skill backend and ask only for permission to start it if needed.\n\n### Step 3 - Prepare environment and secrets\n\nRead `references\u002Fenv-and-secrets.md` before changing `deploy\u002F.env`.\n\n```bash\nif [ ! -f deploy\u002F.env ]; then\n  cp deploy\u002F.env.example deploy\u002F.env\n  echo \"created deploy\u002F.env from deploy\u002F.env.example\"\nfi\n```\n\nExpected output when the file is missing: `created deploy\u002F.env from deploy\u002F.env.example`. Expected output when the file\nalready exists: no output, and the existing file is preserved.\n\nNever print secret values. If credentials are missing, ask the user to update `deploy\u002F.env`; do not ask them to paste\nsecret values into chat.\n\n### Step 4 - Route to the selected deployment path\n\nMatch the user request, then read the referenced file before acting:\n\n| User Intent | Reference |\n|---|---|\n| No AI-Q checkout exists, install AIQ, clone AIQ, locate repo | `references\u002Flocate-or-clone.md` |\n| Configure environment, check API keys, inspect `.env` | `references\u002Fenv-and-secrets.md` |\n| Choose an AI-Q workflow config, understand config files, set `BACKEND_CONFIG` or `CONFIG_FILE` | `references\u002Fconfigs.md` |\n| Backend-only local server for `aiq-research`, AIQ as an Agent Skill | `references\u002Fskill-backend.md` |\n| Terminal assistant, CLI-only run, no web UI | `references\u002Fterminal-cli.md` |\n| Quick local development run, start UI\u002Fbackend without containers | `references\u002Flocal-web.md` |\n| Default durable local deployment, Docker Compose, containers, PostgreSQL | `references\u002Fdocker-compose.md` |\n| Kubernetes, Helm, cluster deployment | `references\u002Fkubernetes-helm.md` |\n| Foundational RAG \u002F FRAG integration | `references\u002Ffrag.md` |\n| Basic health checks, shallow smoke checks, handoff to `aiq-research` | `references\u002Fvalidation.md` |\n| Optional deep research completion validation | `references\u002Fend-to-end-validation.md` |\n| Logs, unhealthy services, port conflicts, config failures | `references\u002Ftroubleshooting.md` |\n| Stop services, restart, rebuild, safe cleanup | `references\u002Fshutdown.md` |\n\n### Step 5 - Validate and hand off\n\nAfter startup, read `references\u002Fvalidation.md` and run the appropriate checks for the selected mode. For the default\nlocal backend, verify health:\n\n```bash\ncurl -sf http:\u002F\u002Flocalhost:8000\u002Fhealth\n```\n\nExpected output: a successful JSON health response or an empty successful response depending on the server build. If the\ncommand fails, read `references\u002Ftroubleshooting.md` and diagnose before claiming the backend is ready.\n\n`aiq-research` needs a reachable AI-Q server URL. If the backend is on the default port, no extra configuration is\nneeded:\n\n```bash\nAIQ_SERVER_URL=http:\u002F\u002Flocalhost:8000\n```\n\nIf the backend runs elsewhere, tell the user to set:\n\n```bash\nexport AIQ_SERVER_URL=\"http:\u002F\u002Flocalhost:\u003CPORT>\"\n```\n\nDo not continue into deep research or deep research completion validation unless the user asks for it or confirms the\npost-deploy validation prompt. This skill's success criterion is a deployed and basically validated server, not report\ngeneration quality.\n\n## Version Compatibility\n\n**IMPORTANT:** This skill is designed for NVIDIA AI-Q Blueprint version 2.1.0.\n\nSemantic Versioning Compatibility Rules:\n\n```text\nSkill version: X.Y.Z\nBlueprint version: A.B.C\n\nCompatible IF:\n1. A == X (Major versions MUST match)\n2. B >= Y (Minor version must be equal or greater)\n3. C can be anything (Patch version does not affect compatibility)\n```\n\nExamples:\n\n- Skill version 2.1.0 is compatible with Blueprint version 2.1.0.\n- Skill version 2.1.0 is compatible with Blueprint version 2.2.0.\n- Skill version 2.1.0 is compatible with Blueprint version 2.1.5.\n- Skill version 2.1.0 is not compatible with Blueprint version 3.0.0.\n- Skill version 2.1.0 is not compatible with Blueprint version 2.0.0.\n\nIf your Blueprint version is not compatible:\n\n1. Check for an updated skill version matching your Blueprint version.\n2. Use a Blueprint version compatible with this skill.\n3. Proceed with caution only when the user accepts the compatibility risk; deployment commands or config names may have\n   changed.\n\n## Security Best Practices\n\n- Never print secret values. Check only whether required environment variables are set.\n- Store credentials in `deploy\u002F.env` or environment variables, not in chat transcripts, shell history, committed files,\n  or example commands.\n- Do not overwrite `deploy\u002F.env` when it already exists.\n- Ask before destructive cleanup such as deleting Docker volumes with `down -v`.\n- Do not claim FRAG is ready unless both `RAG_SERVER_URL` and `RAG_INGEST_URL` are configured and reachable.\n- Run verification commands yourself when possible.\n\n## Limitations\n\n- This skill prepares and validates AI-Q infrastructure; it does not judge deep research report quality.\n- It cannot provide or inspect secret values. Users must configure credentials outside chat.\n- Helm, FRAG, custom config, and self-hosted model paths depend on infrastructure the user controls.\n- Destructive cleanup, such as deleting Docker volumes, requires explicit user approval.\n\n## Examples\n\n### Example 1: Deploy a backend-only Skill server with Docker Compose\n\n```bash\ntest -f deploy\u002F.env || cp deploy\u002F.env.example deploy\u002F.env\ngit check-ignore deploy\u002F.env\ncd deploy\u002Fcompose\nBUILD_TARGET=release docker compose --env-file ..\u002F.env -f docker-compose.yaml config --quiet\nBUILD_TARGET=release docker compose --env-file ..\u002F.env -f docker-compose.yaml up -d --build aiq-agent\ncurl -sf http:\u002F\u002Flocalhost:8000\u002Fhealth\n```\n\nExpected output:\n\n```text\ndeploy\u002F.env\n\u003Cdocker compose starts aiq-agent and dependencies>\n\u003Chealth endpoint returns a successful response>\n```\n\nIf Docker, ports, credentials, or health checks fail, read `references\u002Ftroubleshooting.md` before retrying.\n\n### Example 2: Hand off a non-default backend URL to aiq-research\n\n```bash\nexport AIQ_SERVER_URL=\"http:\u002F\u002Flocalhost:8100\"\ncurl -sf \"$AIQ_SERVER_URL\u002Fhealth\"\n```\n\nExpected output: a successful health response. Then tell the user to keep `AIQ_SERVER_URL` set before invoking\n`aiq-research`.\n\n## References\n\n| Topic | Documentation |\n|---|---|\n| Locate or clone AI-Q | `references\u002Flocate-or-clone.md` |\n| Environment and secrets | `references\u002Fenv-and-secrets.md` |\n| Workflow configs | `references\u002Fconfigs.md` |\n| Agent Skill backend | `references\u002Fskill-backend.md` |\n| CLI deployment | `references\u002Fterminal-cli.md` |\n| Local web deployment | `references\u002Flocal-web.md` |\n| Docker Compose deployment | `references\u002Fdocker-compose.md` |\n| Kubernetes and Helm deployment | `references\u002Fkubernetes-helm.md` |\n| FRAG integration | `references\u002Ffrag.md` |\n| Basic validation | `references\u002Fvalidation.md` |\n| End-to-end validation | `references\u002Fend-to-end-validation.md` |\n| Troubleshooting | `references\u002Ftroubleshooting.md` |\n| Shutdown and cleanup | `references\u002Fshutdown.md` |\n\n## Common Issues\n\n### Issue: Backend port is already in use\n\n**Symptoms:**\n\n- Docker Compose fails to bind port `8000`.\n- `curl -sf http:\u002F\u002Flocalhost:8000\u002Fhealth` reaches an unexpected service or fails.\n\n**Causes:**\n\n- Another AI-Q backend or local development server is already running.\n- `PORT` in `deploy\u002F.env` conflicts with an existing process.\n\n**Solutions:**\n\n1. Identify the process:\n   ```bash\n   lsof -nP -iTCP:8000 -sTCP:LISTEN\n   ```\n2. Either stop the conflicting process with the user's approval or set a different port in `deploy\u002F.env`, such as\n   `PORT=8100`.\n3. Restart the selected deployment path and verify:\n   ```bash\n   curl -sf http:\u002F\u002Flocalhost:8100\u002Fhealth\n   ```\n\n### Issue: Required credentials are missing\n\n**Symptoms:**\n\n- Infrastructure starts, but model-backed chat or research requests fail.\n- Logs mention unauthorized, forbidden, invalid key, or missing provider configuration.\n\n**Causes:**\n\n- `NVIDIA_API_KEY` is missing or empty.\n- No supported search provider key is configured for web research.\n\n**Solutions:**\n\n1. Check presence without printing values by following `references\u002Fenv-and-secrets.md`.\n2. Ask the user to update `deploy\u002F.env`; do not ask them to paste secrets into chat.\n3. Rerun `references\u002Fvalidation.md` after the user updates credentials.\n\n### Issue: Backend is healthy but not compatible with aiq-research\n\n**Symptoms:**\n\n- `\u002Fhealth` succeeds, but `\u002Fchat` or `\u002Fv1\u002Fjobs\u002Fasync\u002Fagents` fails.\n- `aiq-research` reports that async agents are unavailable.\n\n**Causes:**\n\n- The selected config is CLI-only or does not expose the web\u002FAPI backend expected by the skill.\n- `BACKEND_CONFIG` or `CONFIG_FILE` points at the wrong AI-Q config.\n\n**Solutions:**\n\n1. Read `references\u002Fconfigs.md` and confirm the selected config is API-enabled.\n2. For the default Skill backend, use `configs\u002Fconfig_web_default_llamaindex.yml`.\n3. Restart the backend and rerun `references\u002Fvalidation.md`.\n\n### Issue: Docker cleanup would remove useful state\n\n**Symptoms:**\n\n- Troubleshooting suggests `docker compose down -v`.\n- The user may have local PostgreSQL job or checkpoint data they want to keep.\n\n**Causes:**\n\n- `down -v` removes Docker volumes.\n- Rebuilds and restarts are often enough for config or image changes.\n\n**Solutions:**\n\n1. Prefer a normal restart from `references\u002Fshutdown.md`.\n2. Ask for explicit approval before running volume deletion.\n3. After cleanup, rerun deployment and validation from the selected route.\n","# AIQ Deploy Skill\n\n## Purpose\n\nUse this skill to get a local or self-hosted NVIDIA AI-Q Blueprint server running and verified for use by\n`aiq-research`.\n\nThis skill owns setup, deployment, operational checks, troubleshooting, and shutdown. It does not run deep\nresearch itself. After deployment is healthy, hand off the verified server URL to `aiq-research`.\nThe workflow stays explicit so deployment validation and handoff are repeatable across supported agent clients.\n\n## Prerequisites\n\nUsers need:\n\n- Access to clone or update `https:\u002F\u002Fgithub.com\u002FNVIDIA-AI-Blueprints\u002Faiq`.\n- Git available in the shell.\n- One deployment runtime:\n  - Docker Engine with Docker Compose v2 for the default durable local deployment.\n  - Python 3.11+ and `uv` for local process or CLI mode.\n  - Node.js 20+ and `npm` for local browser UI development mode.\n  - `kubectl` 1.28+, Helm 3.12+, and access to a Kubernetes cluster for Helm mode.\n- Network access to GitHub, NVIDIA-hosted model endpoints, and any selected search provider.\n- Credentials stored outside chat. Hosted-model usage requires `NVIDIA_API_KEY`; web research requires at least\n  one supported search provider key such as `TAVILY_API_KEY`, `SERPER_API_KEY`, or `EXA_API_KEY`.\n- System capacity for the selected runtime. Docker Compose mode starts the AI-Q backend and PostgreSQL by default;\n  browser UI mode also uses frontend port `3000`. Self-hosted model or RAG deployments may require GPU resources.\n\nBefore writing secrets, verify `deploy\u002F.env` is ignored:\n\n```bash\ngit check-ignore deploy\u002F.env\n```\n\nExpected output: `deploy\u002F.env` or a matching ignore rule. If it is not ignored, stop and fix the ignore rule before\nplacing credentials in the file.\n\n## Instructions\n\n1. Locate or clone the AI-Q repository.\n2. Confirm the expected repository files exist.\n3. Select the deployment mode.\n4. Prepare `deploy\u002F.env` without overwriting user secrets.\n5. Check runtime prerequisites for the selected path.\n6. Start the selected deployment.\n7. Run basic validation.\n8. Report the verified `AIQ_SERVER_URL` for `aiq-research`.\n9. Ask whether to run optional deep research completion validation.\n\n### Step 1 - Locate or clone AI-Q\n\nIf no AI-Q checkout exists, read `references\u002Flocate-or-clone.md` before cloning. In an existing checkout, confirm the\nrequired files:\n\n```bash\npwd\ntest -f pyproject.toml\ntest -f deploy\u002F.env.example\ntest -d configs\n```\n\nExpected output: `pwd` prints the AI-Q repository path; the `test` commands exit with status 0 and no output.\n\n### Step 2 - Select the deployment mode\n\nIf the user asks to install, deploy, set up, or run AI-Q without naming a mode, ask:\n\n```text\nHow do you want to run AI-Q?\n\n1. Skill backend - backend-only service for aiq-research w\u002Fo browser UI.\n2. CLI - interactive terminal AI-Q.\n3. UI - browser AI-Q app with backend and frontend.\n4. Custom - choose an existing AI-Q config or review advanced customization docs before deployment.\n```\n\nWait for the user's answer before starting services.\n\nDo not ask this question when the user already specified a mode, such as Docker Compose, Helm, UI, CLI, or Agent Skill\nbackend. Do not ask the full mode question when `aiq-research` routed here because a deep research request needs a\nbackend. In that case, prefer Agent Skill backend and ask only for permission to start it if needed.\n\n### Step 3 - Prepare environment and secrets\n\nRead `references\u002Fenv-and-secrets.md` before changing `deploy\u002F.env`.\n\n```bash\nif [ ! -f deploy\u002F.env ]; then\n  cp deploy\u002F.env.example deploy\u002F.env\n  echo \"created deploy\u002F.env from deploy\u002F.env.example\"\nfi\n```\n\nExpected output when the file is missing: `created deploy\u002F.env from deploy\u002F.env.example`. Expected output when the file\nalready exists: no output, and the existing file is preserved.\n\nNever print secret values. If credentials are missing, ask the user to update `deploy\u002F.env`; do not ask them to paste\nsecret values into chat.\n\n### Step 4 - Route to the selected deployment path\n\nMatch the user request, then read the referenced file before acting:\n\n| User Intent | Reference |\n|---|---|\n| No AI-Q checkout exists, install AIQ, clone AIQ, locate repo | `references\u002Flocate-or-clone.md` |\n| Configure environment, check API keys, inspect `.env` | `references\u002Fenv-and-secrets.md` |\n| Choose an AI-Q workflow config, understand config files, set `BACKEND_CONFIG` or `CONFIG_FILE` | `references\u002Fconfigs.md` |\n| Backend-only local server for `aiq-research`, AIQ as an Agent Skill | `references\u002Fskill-backend.md` |\n| Terminal assistant, CLI-only run, no web UI | `references\u002Fterminal-cli.md` |\n| Quick local development run, start UI\u002Fbackend without containers | `references\u002Flocal-web.md` |\n| Default durable local deployment, Docker Compose, containers, PostgreSQL | `references\u002Fdocker-compose.md` |\n| Kubernetes, Helm, cluster deployment | `references\u002Fkubernetes-helm.md` |\n| Foundational RAG \u002F FRAG integration | `references\u002Ffrag.md` |\n| Basic health checks, shallow smoke checks, handoff to `aiq-research` | `references\u002Fvalidation.md` |\n| Optional deep research completion validation | `references\u002Fend-to-end-validation.md` |\n| Logs, unhealthy services, port conflicts, config failures | `references\u002Ftroubleshooting.md` |\n| Stop services, restart, rebuild, safe cleanup | `references\u002Fshutdown.md` |\n\n### Step 5 - Validate and hand off\n\nAfter startup, read `references\u002Fvalidation.md` and run the appropriate checks for the selected mode. For the default\nlocal backend, verify health:\n\n```bash\ncurl -sf http:\u002F\u002Flocalhost:8000\u002Fhealth\n```\n\nExpected output: a successful JSON health response or an empty successful response depending on the server build. If the\ncommand fails, read `references\u002Ftroubleshooting.md` and diagnose before claiming the backend is ready.\n\n`aiq-research` needs a reachable AI-Q server URL. If the backend is on the default port, no extra configuration is\nneeded:\n\n```bash\nAIQ_SERVER_URL=http:\u002F\u002Flocalhost:8000\n```\n\nIf the backend runs elsewhere, tell the user to set:\n\n```bash\nexport AIQ_SERVER_URL=\"http:\u002F\u002Flocalhost:\u003CPORT>\"\n```\n\nDo not continue into deep research or deep research completion validation unless the user asks for it or confirms the\npost-deploy validation prompt. This skill's success criterion is a deployed and basically validated server, not report\ngeneration quality.\n\n## Version Compatibility\n\n**IMPORTANT:** This skill is designed for NVIDIA AI-Q Blueprint version 2.1.0.\n\nSemantic Versioning Compatibility Rules:\n\n```text\nSkill version: X.Y.Z\nBlueprint version: A.B.C\n\nCompatible IF:\n1. A == X (Major versions MUST match)\n2. B >= Y (Minor version must be equal or greater)\n3. C can be anything (Patch version does not affect compatibility)\n```\n\nExamples:\n\n- Skill version 2.1.0 is compatible with Blueprint version 2.1.0.\n- Skill version 2.1.0 is compatible with Blueprint version 2.2.0.\n- Skill version 2.1.0 is compatible with Blueprint version 2.1.5.\n- Skill version 2.1.0 is not compatible with Blueprint version 3.0.0.\n- Skill version 2.1.0 is not compatible with Blueprint version 2.0.0.\n\nIf your Blueprint version is not compatible:\n\n1. Check for an updated skill version matching your Blueprint version.\n2. Use a Blueprint version compatible with this skill.\n3. Proceed with caution only when the user accepts the compatibility risk; deployment commands or config names may have\n   changed.\n\n## Security Best Practices\n\n- Never print secret values. Check only whether required environment variables are set.\n- Store credentials in `deploy\u002F.env` or environment variables, not in chat transcripts, shell history, committed files,\n  or example commands.\n- Do not overwrite `deploy\u002F.env` when it already exists.\n- Ask before destructive cleanup such as deleting Docker volumes with `down -v`.\n- Do not claim FRAG is ready unless both `RAG_SERVER_URL` and `RAG_INGEST_URL` are configured and reachable.\n- Run verification commands yourself when possible.\n\n## Limitations\n\n- This skill prepares and validates AI-Q infrastructure; it does not judge deep research report quality.\n- It cannot provide or inspect secret values. Users must configure credentials outside chat.\n- Helm, FRAG, custom config, and self-hosted model paths depend on infrastructure the user controls.\n- Destructive cleanup, such as deleting Docker volumes, requires explicit user approval.\n\n## Examples\n\n### Example 1: Deploy a backend-only Skill server with Docker Compose\n\n```bash\ntest -f deploy\u002F.env || cp deploy\u002F.env.example deploy\u002F.env\ngit check-ignore deploy\u002F.env\ncd deploy\u002Fcompose\nBUILD_TARGET=release docker compose --env-file ..\u002F.env -f docker-compose.yaml config --quiet\nBUILD_TARGET=release docker compose --env-file ..\u002F.env -f docker-compose.yaml up -d --build aiq-agent\ncurl -sf http:\u002F\u002Flocalhost:8000\u002Fhealth\n```\n\nExpected output:\n\n```text\ndeploy\u002F.env\n\u003Cdocker compose starts aiq-agent and dependencies>\n\u003Chealth endpoint returns a successful response>\n```\n\nIf Docker, ports, credentials, or health checks fail, read `references\u002Ftroubleshooting.md` before retrying.\n\n### Example 2: Hand off a non-default backend URL to aiq-research\n\n```bash\nexport AIQ_SERVER_URL=\"http:\u002F\u002Flocalhost:8100\"\ncurl -sf \"$AIQ_SERVER_URL\u002Fhealth\"\n```\n\nExpected output: a successful health response. Then tell the user to keep `AIQ_SERVER_URL` set before invoking\n`aiq-research`.\n\n## References\n\n| Topic | Documentation |\n|---|---|\n| Locate or clone AI-Q | `references\u002Flocate-or-clone.md` |\n| Environment and secrets | `references\u002Fenv-and-secrets.md` |\n| Workflow configs | `references\u002Fconfigs.md` |\n| Agent Skill backend | `references\u002Fskill-backend.md` |\n| CLI deployment | `references\u002Fterminal-cli.md` |\n| Local web deployment | `references\u002Flocal-web.md` |\n| Docker Compose deployment | `references\u002Fdocker-compose.md` |\n| Kubernetes and Helm deployment | `references\u002Fkubernetes-helm.md` |\n| FRAG integration | `references\u002Ffrag.md` |\n| Basic validation | `references\u002Fvalidation.md` |\n| End-to-end validation | `references\u002Fend-to-end-validation.md` |\n| Troubleshooting | `references\u002Ftroubleshooting.md` |\n| Shutdown and cleanup | `references\u002Fshutdown.md` |\n\n## Common Issues\n\n### Issue: Backend port is already in use\n\n**Symptoms:**\n\n- Docker Compose fails to bind port `8000`.\n- `curl -sf http:\u002F\u002Flocalhost:8000\u002Fhealth` reaches an unexpected service or fails.\n\n**Causes:**\n\n- Another AI-Q backend or local development server is already running.\n- `PORT` in `deploy\u002F.env` conflicts with an existing process.\n\n**Solutions:**\n\n1. Identify the process:\n   ```bash\n   lsof -nP -iTCP:8000 -sTCP:LISTEN\n   ```\n2. Either stop the conflicting process with the user's approval or set a different port in `deploy\u002F.env`, such as\n   `PORT=8100`.\n3. Restart the selected deployment path and verify:\n   ```bash\n   curl -sf http:\u002F\u002Flocalhost:8100\u002Fhealth\n   ```\n\n### Issue: Required credentials are missing\n\n**Symptoms:**\n\n- Infrastructure starts, but model-backed chat or research requests fail.\n- Logs mention unauthorized, forbidden, invalid key, or missing provider configuration.\n\n**Causes:**\n\n- `NVIDIA_API_KEY` is missing or empty.\n- No supported search provider key is configured for web research.\n\n**Solutions:**\n\n1. Check presence without printing values by following `references\u002Fenv-and-secrets.md`.\n2. Ask the user to update `deploy\u002F.env`; do not ask them to paste secrets into chat.\n3. Rerun `references\u002Fvalidation.md` after the user updates credentials.\n\n### Issue: Backend is healthy but not compatible with aiq-research\n\n**Symptoms:**\n\n- `\u002Fhealth` succeeds, but `\u002Fchat` or `\u002Fv1\u002Fjobs\u002Fasync\u002Fagents` fails.\n- `aiq-research` reports that async agents are unavailable.\n\n**Causes:**\n\n- The selected config is CLI-only or does not expose the web\u002FAPI backend expected by the skill.\n- `BACKEND_CONFIG` or `CONFIG_FILE` points at the wrong AI-Q config.\n\n**Solutions:**\n\n1. Read `references\u002Fconfigs.md` and confirm the selected config is API-enabled.\n2. For the default Skill backend, use `configs\u002Fconfig_web_default_llamaindex.yml`.\n3. Restart the backend and rerun `references\u002Fvalidation.md`.\n\n### Issue: Docker cleanup would remove useful state\n\n**Symptoms:**\n\n- Troubleshooting suggests `docker compose down -v`.\n- The user may have local PostgreSQL job or checkpoint data they want to keep.\n\n**Causes:**\n\n- `down -v` removes Docker volumes.\n- Rebuilds and restarts are often enough for config or image changes.\n\n**Solutions:**\n\n1. Prefer a normal restart from `references\u002Fshutdown.md`.\n2. Ask for explicit approval before running volume deletion.\n3. After cleanup, rerun deployment and validation from the selected route.\n",{"name":5,"license":15,"metadata":24,"description":35,"allowed-tools":36,"compatibility":37},{"tags":25,"author":32,"version":33,"github-url":34},[26,27,28,29,30,31],"nvidia","aiq","blueprint","deploy","operations","agent-skills","NVIDIA AI-Q Blueprint Team \u003Caiq-blueprint@nvidia.com>","2.1.0","https:\u002F\u002Fgithub.com\u002FNVIDIA-AI-Blueprints\u002Faiq","Use when asked to install, deploy, run, validate, troubleshoot, or stop NVIDIA AI-Q Blueprint infrastructure.\n","Read Bash","Designed for Claude Code, OpenCode, Codex, and Agent Skills-compatible tools. Requires Git, network\naccess to GitHub, and one selected runtime path: Docker Compose v2 for the default local deployment,\nPython 3.11+ and uv for local process or CLI mode, Node.js 20+ and npm for local web UI mode, or\nkubectl 1.28+ and Helm 3.12+ for Kubernetes and Helm mode.\n","main","e3091d1b371a9dcd9455d333175f948b286fdfcf","plugins\u002Fnvidia\u002Fskills\u002Faiq-deploy",1,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh","https:\u002F\u002Fkael-odin.github.io\u002Fcodex-plugins-zh\u002F","https:\u002F\u002Fgithub.com\u002Fkael-odin","https:\u002F\u002Fkael-odin.github.io\u002Fodin-valhalla\u002F","Full-stack development • Open-source • System engineering\r\nTurning ideas into repositories.","2026-11-08 17:05:36",{"source":8,"skillName":5,"commands":49,"prompt":59,"zipAvailable":60},[50,53,56],{"tool":51,"command":52},"npx","npx skills add kael-odin\u002Fcodex-plugins-zh --skill aiq-deploy",{"tool":54,"command":55},"bunx","bunx skills add kael-odin\u002Fcodex-plugins-zh --skill aiq-deploy",{"tool":57,"command":58},"pnpm","pnpm dlx skills add kael-odin\u002Fcodex-plugins-zh --skill aiq-deploy","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002FSKILL.md",true,[62,68,72,76,80,84,88,92,96,100,104,108,112,116,120,124,128,132],{"path":63,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":66,"sourceHtmlUrl":67},"agents\u002Fopenai.yaml","REFERENCE","参考文档",173,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Fagents\u002Fopenai.yaml",{"path":69,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":70,"sourceHtmlUrl":71},"BENCHMARK.md",4173,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002FBENCHMARK.md",{"path":73,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":74,"sourceHtmlUrl":75},"evals\u002Fevals.json",1405,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Fevals\u002Fevals.json",{"path":77,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":78,"sourceHtmlUrl":79},"references\u002Fconfigs.md",3539,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Freferences\u002Fconfigs.md",{"path":81,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":82,"sourceHtmlUrl":83},"references\u002Fdocker-compose.md",3550,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Freferences\u002Fdocker-compose.md",{"path":85,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":86,"sourceHtmlUrl":87},"references\u002Fend-to-end-validation.md",5737,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Freferences\u002Fend-to-end-validation.md",{"path":89,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":90,"sourceHtmlUrl":91},"references\u002Fenv-and-secrets.md",3880,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Freferences\u002Fenv-and-secrets.md",{"path":93,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":94,"sourceHtmlUrl":95},"references\u002Ffrag.md",1679,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Freferences\u002Ffrag.md",{"path":97,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":98,"sourceHtmlUrl":99},"references\u002Fkubernetes-helm.md",983,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Freferences\u002Fkubernetes-helm.md",{"path":101,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":102,"sourceHtmlUrl":103},"references\u002Flocal-web.md",1234,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Freferences\u002Flocal-web.md",{"path":105,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":106,"sourceHtmlUrl":107},"references\u002Flocate-or-clone.md",1300,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Freferences\u002Flocate-or-clone.md",{"path":109,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":110,"sourceHtmlUrl":111},"references\u002Fshutdown.md",2214,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Freferences\u002Fshutdown.md",{"path":113,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":114,"sourceHtmlUrl":115},"references\u002Fskill-backend.md",1457,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Freferences\u002Fskill-backend.md",{"path":117,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":118,"sourceHtmlUrl":119},"references\u002Fterminal-cli.md",755,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Freferences\u002Fterminal-cli.md",{"path":121,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":122,"sourceHtmlUrl":123},"references\u002Ftroubleshooting.md",1247,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Freferences\u002Ftroubleshooting.md",{"path":125,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":126,"sourceHtmlUrl":127},"references\u002Fvalidation.md",3229,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Freferences\u002Fvalidation.md",{"path":129,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":130,"sourceHtmlUrl":131},"skill-card.md",4067,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Fskill-card.md",{"path":133,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":136,"sourceHtmlUrl":137},"skill.oms.sig","OTHER","其他",7685,"https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fnvidia\u002Fskills\u002Faiq-deploy\u002Fskill.oms.sig",[139,147,155,163,170,177],{"skillNo":140,"name":141,"nameZh":6,"description":142,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":143,"sourceHtmlUrl":144,"categories":145,"keywords":146,"publishedAt":19},"01M7EDVCKB89EF755PQMKV1BZA","android-emulator-qa","Use when validating Android feature flows in an emulator with adb-driven launch, input, UI-tree inspection, screenshots, and logcat capture.","https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Ftest-android-apps\u002Fskills\u002Fandroid-emulator-qa\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Ftest-android-apps\u002Fskills\u002Fandroid-emulator-qa\u002FSKILL.md",[],[],{"skillNo":148,"name":149,"nameZh":6,"description":150,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":151,"licenseFileUrl":6,"sourceHtmlUrl":152,"categories":153,"keywords":154,"publishedAt":19},"01M7EDPX7WHN6HQZ0QKWB0BGVZ","supabase-postgres-best-practices","Postgres performance optimization and best practices from Supabase. Use this skill when writing, reviewing, or optimizing Postgres queries, schema designs, or database configurations.","MIT","https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fsupabase\u002Fskills\u002Fsupabase-postgres-best-practices\u002FSKILL.md",[],[],{"skillNo":156,"name":157,"nameZh":6,"description":158,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":159,"sourceHtmlUrl":160,"categories":161,"keywords":162,"publishedAt":19},"01M7EDHHD98T8ZH09F40Q973G3","sentry","Use when the user asks to inspect Sentry issues or events, summarize recent production errors, or pull basic Sentry health data via the Sentry API; perform read-only queries with the bundled script and require `SENTRY_AUTH_TOKEN`.","https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fsentry\u002Fskills\u002Fsentry\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Fsentry\u002Fskills\u002Fsentry\u002FSKILL.md",[],[],{"skillNo":164,"name":165,"nameZh":6,"description":166,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":151,"licenseFileUrl":6,"sourceHtmlUrl":167,"categories":168,"keywords":169,"publishedAt":19},"01M7EDH3SGVMS1AFS6Q40B5J48","render-workflows","搭建、开发、测试和部署 Render Workflows。涵盖首次脚手架搭建（通过 CLI 或手动方式）、SDK 安装（Python 或 TypeScript）、任务模式（重试、子任务、扇出）、本地开发、控制台部署和故障排查。适用于用户首次搭建 Render Workflows、构建工作流服务、添加或修改工作流任务、本地测试工作流或部署到 Render 的场景。","https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Frender\u002Fskills\u002Frender-workflows\u002FSKILL.md",[],[],{"skillNo":171,"name":172,"nameZh":6,"description":173,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":151,"licenseFileUrl":6,"sourceHtmlUrl":174,"categories":175,"keywords":176,"publishedAt":19},"01M7EDH0CYEKQ4VVARHZAQ43S8","render-web-services","配置 Render web service——端口绑定、TLS、健康检查、自定义域名、自动部署、PR 预览、持久化磁盘及部署生命周期。当用户需要搭建 web service、修复健康检查失败、添加自定义域名、配置零停机部署，或排查端口绑定问题时使用。","https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Frender\u002Fskills\u002Frender-web-services\u002FSKILL.md",[],[],{"skillNo":178,"name":179,"nameZh":6,"description":180,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":151,"licenseFileUrl":6,"sourceHtmlUrl":181,"categories":182,"keywords":183,"publishedAt":19},"01M7EDGVDYFBG1CKSNMQ2DD336","render-static-sites","在 Render 全球 CDN 上部署并配置静态站点——构建命令、发布路径、SPA 路由、重定向、自定义 header 及 PR 预览。当用户需要部署静态站点、搭建 React\u002FVue\u002FHugo\u002FGatsby 前端、配置 SPA 回退路由、添加重定向规则、自定义响应 header，或在静态站点与 web service 之间为前端做选择时使用。触发词:static site、CDN、SPA、single-page app、React deploy、Vue deploy、Hugo、Gatsby、Docusaurus、Jekyll、staticPublishPath。","https:\u002F\u002Fgithub.com\u002Fkael-odin\u002Fcodex-plugins-zh\u002Fblob\u002Fmain\u002Fplugins\u002Frender\u002Fskills\u002Frender-static-sites\u002FSKILL.md",[],[]]