NO-ISSUE: chore(preset): standardize preset naming and enforce AGENTS.md compliance

[hhk7734]

No description provided.

[Copilot]

Pull request overview

Standardizes Odin preset naming/labeling and improves template robustness by switching runtime-base and utility templates to nil-safe parallelism field access, while updating docs and helm contribution guidance to align with the new conventions.

Changes:

• Update docs (Quickstart + Odin skill) to reference standardized preset names and document nil-safe template patterns.
• Switch runtime-base and utility templates from direct .Spec.Parallelism.* access to deref ... | default ... to avoid nil-pointer template panics.
• Rename DeepSeek R1 preset template resource names/labels and add preset naming guidance to deploy/helm/AGENTS.md.

Reviewed changes

Copilot reviewed 19 out of 19 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
website/docs/getting-started/quickstart.mdx	Updates quickstart preset reference and link target; keeps TP2 configured via preset.
skills/guide-odin/SKILL.md	Improves guide formatting and documents deref for nil-safe template access.
deploy/helm/moai-inference-preset/templates/utils/sim-prefill-dp.helm.yaml	Uses deref+default for DP sizing to avoid nil .Spec.Parallelism access.
deploy/helm/moai-inference-preset/templates/utils/sim-dp.helm.yaml	Same nil-safe DP sizing updates.
deploy/helm/moai-inference-preset/templates/utils/sim-decode-dp.helm.yaml	Aligns proxy startup with multi-proxy-per-rank approach and nil-safe DP sizing.
deploy/helm/moai-inference-preset/templates/runtime-bases/vllm.helm.yaml	Nil-safe tensor/expert parallelism rendering via deref.
deploy/helm/moai-inference-preset/templates/runtime-bases/vllm-prefill.helm.yaml	Nil-safe tensor/expert parallelism rendering via deref.
deploy/helm/moai-inference-preset/templates/runtime-bases/vllm-prefill-pp.helm.yaml	Nil-safe pipeline/tensor/expert parallelism rendering via deref.
deploy/helm/moai-inference-preset/templates/runtime-bases/vllm-prefill-dp.helm.yaml	Nil-safe DP sizing + tensor/expert parallelism rendering via deref.
deploy/helm/moai-inference-preset/templates/runtime-bases/vllm-pp.helm.yaml	Nil-safe pipeline/tensor/expert parallelism rendering via deref.
deploy/helm/moai-inference-preset/templates/runtime-bases/vllm-dp.helm.yaml	Nil-safe DP sizing + tensor/expert parallelism rendering via deref.
deploy/helm/moai-inference-preset/templates/runtime-bases/vllm-decode.helm.yaml	Nil-safe tensor/expert parallelism rendering via deref.
deploy/helm/moai-inference-preset/templates/runtime-bases/vllm-decode-pp.helm.yaml	Nil-safe pipeline/tensor/expert parallelism rendering via deref.
deploy/helm/moai-inference-preset/templates/runtime-bases/vllm-decode-dp.helm.yaml	Nil-safe DP sizing + tensor/expert parallelism rendering via deref.
deploy/helm/moai-inference-preset/templates/presets/quickstart/quickstart-vllm-deepseek-ai-deepseek-r1-prefill-amd-mi300x-dp8-moe-ep8.helm.yaml	Renames quickstart DeepSeek preset resource name/labels to standardized parallelism naming.
deploy/helm/moai-inference-preset/templates/presets/quickstart/quickstart-vllm-deepseek-ai-deepseek-r1-decode-amd-mi300x-dp8-moe-ep8.helm.yaml	Same standardized rename for decode preset.
deploy/helm/moai-inference-preset/templates/presets/deepseek-r1/vllm-deepseek-r1-prefill-mi300x-dp8-moe-ep8.helm.yaml	Renames DeepSeek R1 preset resource name/labels (still needs to align with new convention).
deploy/helm/moai-inference-preset/templates/presets/deepseek-r1/vllm-deepseek-r1-decode-mi300x-dp8-moe-ep8.helm.yaml	Renames DeepSeek R1 decode preset resource name/labels (still needs to align with new convention).
deploy/helm/AGENTS.md	Adds explicit Odin preset naming convention, reserved labels, and responsibility boundaries.

Comments suppressed due to low confidence (2)

deploy/helm/moai-inference-preset/templates/presets/deepseek-r1/vllm-deepseek-r1-prefill-mi300x-dp8-moe-ep8.helm.yaml:4

• metadata.name doesn’t follow the preset naming convention added in deploy/helm/AGENTS.md (pattern includes -{accelerator_vendor}-{accelerator_model}-…). This name omits the accelerator vendor (e.g., -amd-), which undermines the new standardization and can cause ambiguity across vendors.
  deploy/helm/moai-inference-preset/templates/presets/deepseek-r1/vllm-deepseek-r1-decode-mi300x-dp8-moe-ep8.helm.yaml:4
• metadata.name doesn’t follow the preset naming convention added in deploy/helm/AGENTS.md (pattern includes -{accelerator_vendor}-{accelerator_model}-…). This name omits the accelerator vendor (e.g., -amd-), which undermines the new standardization and can cause ambiguity across vendors.