Add vLLM server intropsection endpoints

.github/CODEOWNERS

@@ -0,0 +1,5 @@
+# Global owners for the entire repo
+* @hickeyma @vMaroon @elevran
+
+# vLLM extensions
+/vllm/ @hickeyma

.github/ISSUE_TEMPLATE/bug.yaml

@@ -0,0 +1,62 @@
+name: Bug Report
+description: Create a report to help us improve
+type: bug
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Thanks for taking the time to fill out this bug report!
+
+        Please do **not** use this form to report a security vulnerability —
+        follow [SECURITY.md](https://github.com/llm-d/llm-d-api-extensions/blob/main/SECURITY.md) instead.
+  - type: dropdown
+    id: component
+    attributes:
+      label: Component
+      description: First, tell us which component we should focus on
+      options:
+        - I don't know
+        - vLLM / server introspection - config
+        - vLLM / server introspection - devices
+        - vLLM / server introspection - kv-cache
+        - Packaging and install
+        - Documentation
+        - CI
+      default: 0
+    validations:
+      required: true
+  - type: textarea
+    id: describe
+    attributes:
+      label: Describe the bug
+      description: A clear and concise description of what the bug is
+      placeholder: Tell us what's wrong
+    validations:
+      required: true
+  - type: textarea
+    id: reproducer
+    attributes:
+      label: Steps to reproduce
+      description: Steps to reproduce the behavior
+      placeholder: I did (A), then (B) and then I saw (C) error
+    validations:
+      required: true
+  - type: textarea
+    id: environment
+    attributes:
+      label: Environment
+      description: How the server was launched and what it is running
+      placeholder: |
+        - Inference server and version (e.g. vLLM 0.11.0)
+        - llm-d-api-extensions-vllm version
+        - VLLM_PLUGINS value and any relevant `vllm serve` flags
+        - Python version, platform, accelerator
+      render: markdown
+    validations:
+      required: true
+  - type: textarea
+    id: extra
+    attributes:
+      label: Additional context or screenshots
+      description: Add any other context about the problem here
+      placeholder: Anything else you want to say to the report, attach screenshots, this is the place

.github/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,8 @@
+blank_issues_enabled: true
+contact_links:
+  - name: Security vulnerability
+    url: https://github.com/llm-d/llm-d-api-extensions/blob/main/SECURITY.md
+    about: Please report vulnerabilities privately, not as a public issue.
+  - name: llm-d project discussion
+    url: https://github.com/llm-d/llm-d/issues
+    about: For questions about llm-d as a whole rather than these extensions.

.github/ISSUE_TEMPLATE/feature.yaml

@@ -0,0 +1,41 @@
+name: Feature Request
+description: Suggest a new extension, endpoint or improvement
+type: feature
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Thanks for taking the time to suggest an improvement!
+  - type: textarea
+    id: problem
+    attributes:
+      label: What problem does this solve?
+      description: Describe the use case. What does llm-d need that it cannot get today?
+      placeholder: Routing decisions need X, which is not exposed by any current endpoint
+    validations:
+      required: true
+  - type: textarea
+    id: proposal
+    attributes:
+      label: Proposed solution
+      description: |
+        If you are proposing an endpoint, sketch the route and an example response.
+        Note whether it needs an engine, and any server flags it would require.
+    validations:
+      required: true
+  - type: dropdown
+    id: server
+    attributes:
+      label: Inference server
+      options:
+        - vLLM
+        - Another server (please describe below)
+        - Not server specific
+      default: 0
+    validations:
+      required: true
+  - type: textarea
+    id: alternatives
+    attributes:
+      label: Alternatives considered
+      description: Including whether this belongs upstream in the inference server instead

.github/PULL_REQUEST_TEMPLATE/pull_request_template.md

@@ -0,0 +1,39 @@
+---
+name: Pull request
+about: Create a pull request
+---
+
+## Description
+
+Please include a summary of the change and which issue is fixed. Please also include relevant motivation and context. List any dependencies that are required for this change.
+
+Fixes # (issue)
+
+## Type of change
+
+Please delete options that are not relevant.
+
+- [ ] Bug fix (non-breaking change which fixes an issue)
+- [ ] New feature (non-breaking change which adds functionality)
+- [ ] Breaking change (fix or feature that would cause existing functionality to not work as expected)
+- [ ] This change requires a documentation update
+
+## How Has This Been Tested?
+
+Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce.
+
+### Test configuration
+
+- Inference server and version:
+- Python version and platform:
+- `VLLM_PLUGINS` value and relevant `vllm serve` flags:
+
+## Checklist
+
+- [ ] My changes follow the style guidelines of this project
+- [ ] I have performed a self-review of my own changes
+- [ ] I have signed off my commits (`git commit -s`, per the [DCO](https://developercertificate.org/))
+- [ ] `ruff check .` and `ruff format --check .` pass
+- [ ] `pytest tests/ -v` passes
+- [ ] Any new endpoint is opt-in, namespaced under `/plugins/llm-d-<extension>/...` and documented in `vllm/README.md`
+- [ ] I have updated the documentation accordingly

.github/dependabot.yml

@@ -0,0 +1,26 @@
+# Dependabot configuration for llm-d-api-extensions
+# Based on the canonical config from llm-d/llm-d-infra
+
+version: 2
+updates:
+
+  # GitHub Actions dependencies
+  - package-ecosystem: "github-actions"
+    directory: "/"
+    schedule:
+      interval: "weekly"
+    labels:
+      - "dependencies"
+      - "release-note-none"
+    commit-message:
+      prefix: "deps(actions)"
+
+  # Python dependencies for the vLLM extensions
+  - package-ecosystem: "pip"
+    directory: "/vllm"
+    schedule:
+      interval: "weekly"
+    labels:
+      - "dependencies"
+    commit-message:
+      prefix: "deps(pip)"

.github/workflows/ci-lint.yaml

@@ -0,0 +1,37 @@
+name: CI - Lint and Format
+
+on:
+  pull_request:
+    branches:
+      - main
+  push:
+    branches:
+      - main
+
+permissions:
+  contents: read
+
+jobs:
+  ruff:
+    runs-on: ubuntu-latest
+    defaults:
+      run:
+        working-directory: vllm
+    steps:
+      - name: Checkout source
+        uses: actions/checkout@v7
+
+      # Checks only, never modifies files in CI.
+      - name: Run Ruff lint check
+        uses: astral-sh/ruff-action@v3
+        with:
+          version: "0.15.11"
+          src: "./vllm"
+          args: "check --output-format=full"
+
+      - name: Run Ruff format check
+        uses: astral-sh/ruff-action@v3
+        with:
+          version: "0.15.11"
+          src: "./vllm"
+          args: "format --check --diff"

.github/workflows/ci-test.yaml

@@ -0,0 +1,47 @@
+name: CI - Unit Tests
+
+on:
+  pull_request:
+    branches:
+      - main
+  push:
+    branches:
+      - main
+
+permissions:
+  contents: read
+
+jobs:
+  unit-tests:
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        python-version: ["3.10", "3.11", "3.12"]
+    defaults:
+      run:
+        working-directory: vllm
+    steps:
+      - name: Checkout source
+        uses: actions/checkout@v7
+
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+          cache: pip
+
+      # `vllm` is a heavyweight GPU oriented dependency and the unit tests only
+      # need a handful of symbols from it (`vllm.logger`, `vllm.tasks`), so
+      # install it on its own rather than pulling the whole project dependency
+      # set through `pip install -e .`.
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install --extra-index-url https://download.pytorch.org/whl/cpu vllm
+          pip install fastapi 'pydantic>=2' pytest pytest-asyncio httpx
+
+      # The opt-in e2e tests skip themselves unless RUN_VLLM_E2E is set. They
+      # download model weights and launch a real server which CI does not do.
+      - name: Run unit tests
+        run: pytest tests/ -v

.gitignore

@@ -0,0 +1,41 @@
+# Python bytecode
+__pycache__/
+*.py[cod]
+*$py.class
+
+# Packaging / build artifacts
+build/
+dist/
+*.egg-info/
+*.egg
+.eggs/
+wheels/
+
+# Virtual environments
+.venv/
+venv/
+ENV/
+
+# Test and tool caches
+.pytest_cache/
+.mypy_cache/
+.ruff_cache/
+.coverage
+.coverage.*
+htmlcov/
+coverage.xml
+.tox/
+.nox/
+
+# Editors / IDEs
+.idea/
+*.swp
+*.swo
+.vscode/
+
+# OS
+.DS_Store
+Thumbs.db
+
+# Local model / server artifacts pulled down by the opt-in e2e tests
+*.log

CODE_OF_CONDUCT.md

@@ -0,0 +1,34 @@
+## Code of Conduct and Covenant
+
+## Our Pledge
+
+In the interest of fostering an open and welcoming environment, we as contributors and maintainers pledge to making participation in our project and our community a harassment-free experience for everyone, regardless of age, body size, disability, ethnicity, sex characteristics, gender identity and expression, level of experience, education, socio-economic status, nationality, personal appearance, race, religion, or sexual identity and orientation.
+
+## Our Standards
+
+Examples of behavior that contributes to creating a positive environment include:
+
+* Using welcoming and inclusive language
+* Being respectful of differing viewpoints and experiences
+* Gracefully accepting constructive criticism
+* Focusing on what is best for the community
+* Showing empathy towards other community members
+
+Examples of unacceptable behavior by participants include:
+
+* The use of sexualized language or imagery and unwelcome sexual attention or advances
+* Trolling, insulting/derogatory comments, and personal or political attacks
+* Public or private harassment
+* Publishing others' private information, such as a physical or electronic address, without explicit permission
+* Other conduct which could reasonably be considered inappropriate in a professional setting
+
+## Our Responsibilities
+
+Project maintainers have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, or to ban temporarily or permanently any contributor for other behaviors that they deem inappropriate, threatening, offensive, or harmful.
+
+## Scope
+
+This Code of Conduct applies both within project spaces and in public spaces when an individual is representing the project or its community. Examples of representing a project or community include using an official project e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event. Representation of a project may be further defined and clarified by project maintainers.Attribution
+This Code of Conduct is adapted from the Contributor Covenant, version 1.4, available at <https://www.contributor-covenant.org/version/1/4/code-of-conduct.html>
+
+For answers to common questions about this code of conduct, see <https://www.contributor-covenant.org/faq>