Merge pull request #1565 from syedriko/syedriko-ols-1958

LCORE-1958: Correct OpenAPI spec violations in docs/openapi.json, add OpenAPI linting to CI

Author: tisnik

.github/workflows/openapi_spectral.yaml

@@ -0,0 +1,42 @@
+# OpenAPI: regenerate check (Spectral + committed docs/openapi.json drift).
+# - scripts/generate_openapi_schema.py builds the spec from FastAPI (app.main).
+# - CI fails if docs/openapi.json does not match generator output (run locally:
+#   uv run scripts/generate_openapi_schema.py docs/openapi.json).
+name: OpenAPI (Spectral)
+
+on:
+  - push
+  - pull_request
+
+jobs:
+  spectral:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      pull-requests: read
+    steps:
+      - uses: actions/checkout@v4
+      - name: Install uv
+        uses: astral-sh/setup-uv@v5
+        with:
+          python-version: "3.12"
+      - name: Install dependencies
+        # Same pattern as local dev (CONTRIBUTING.md): dev + llslibdev for a full app import.
+        run: uv sync --group dev --group llslibdev
+      - name: Install PDM
+        # scripts/generate_openapi_schema.py asserts OpenAPI info.version matches `pdm show --version`.
+        run: uv pip install pdm
+      - name: Verify docs/openapi.json matches generator
+        run: |
+          set -euo pipefail
+          uv run python scripts/generate_openapi_schema.py /tmp/openapi-generated.json
+          if ! diff -u docs/openapi.json /tmp/openapi-generated.json; then
+            echo "::error::docs/openapi.json is out of date. Regenerate with: uv run scripts/generate_openapi_schema.py docs/openapi.json"
+            exit 1
+          fi
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: "22"
+      - name: Spectral lint
+        run: npx --yes @stoplight/spectral-cli@6 lint docs/openapi.json --fail-severity error --display-only-failures

.spectral.yaml

@@ -0,0 +1,9 @@
+# Spectral (OpenAPI) — https://meta.stoplight.io/docs/spectral/
+#
+# Full Stoplight OAS ruleset. `oas3-valid-media-example` is off: examples in
+# docs/openapi.json are often partial and produced ~600 warnings with little
+# signal until examples are aligned with schemas. Re-enable when tightening
+# examples (set to "warn" or "error").
+extends: spectral:oas
+rules:
+  oas3-valid-media-example: off

CONTRIBUTING.md

@@ -18,6 +18,7 @@
     * [Pre-commit hook settings](#pre-commit-hook-settings)
 * [Code coverage measurement](#code-coverage-measurement)
 * [Linters](#linters)
+    * [OpenAPI (Spectral)](#openapi-spectral)
     * [Type hints checks](#type-hints-checks)
     * [Ruff](#ruff)
     * [Pylint](#pylint)
@@ -48,6 +49,7 @@
 - git
 - Python 3.12 or 3.13
 - pip
+- **Node.js** (18 or newer; **npm** and **`npx`** ship with Node). **`make verify`** runs **`lint-openapi`**, which calls **`npx --yes @stoplight/spectral-cli@6`** (see `Makefile`). If **`npx`** is not installed, **`lint-openapi` is skipped** with a message so **`make verify` still succeeds** locally; install Node to run the OpenAPI check. **CI** always runs Spectral (see `.github/workflows/openapi_spectral.yaml`).
 
 The development requires at least [Python 3.12](https://docs.python.org/3/whatsnew/3.12.html) due to significant improvement on performance, optimizations which benefit modern ML, AI, LLM, NL stacks, and improved asynchronous processing capabilities. It is also possible to use Python 3.13.
 
@@ -57,6 +59,7 @@ The development requires at least [Python 3.12](https://docs.python.org/3/whatsn
 
 1. `pip install --user uv`
 1. `uv --version` -- should return no error
+1. Install [Node.js](https://nodejs.org/en/download) (LTS is fine) or use your OS package manager, e.g. Fedora: `sudo dnf install nodejs`, macOS with [Homebrew](https://brew.sh/): `brew install node`. Confirm `node --version` and `npx --version` work. CI uses Node 22 for Spectral (see `.github/workflows/openapi_spectral.yaml`).
 
 
 
@@ -217,6 +220,10 @@ Code coverage reports are generated in JSON and also in format compatible with [
 
 _Black_, _Ruff_, Pyright, _Pylint_, __Pydocstyle__, __Mypy__, and __Bandit__ tools are used as linters. There are a bunch of linter rules enabled for this repository. All of them are specified in `pyproject.toml`, such as in sections `[tool.ruff]` and `[tool.pylint."MESSAGES CONTROL"]`. Some specific rules can be disabled using `ignore` parameter (empty now).
 
+### OpenAPI (Spectral)
+
+OpenAPI is linted with [Spectral](https://stoplight.io/open-api/) via **`npx --yes @stoplight/spectral-cli@6`** in the **`lint-openapi`** target (`make lint-openapi`, part of **`make verify`**). If **`npx`** is missing, **`lint-openapi`** skips Spectral locally; install **Node.js** to run it (see [Prerequisites](#prerequisites) and [Tooling installation](#tooling-installation)). **CI** always runs the Spectral step. If you introduce a **new** router tag (`APIRouter(tags=[...])`), you must also extend the global tag list in `src/app/main.py` and regenerate `docs/openapi.json`. See **[docs/contributing/openapi-tags-and-spectral.md](docs/contributing/openapi-tags-and-spectral.md)**.
+
 
 ### Type hints checks
 

Makefile

@@ -113,13 +113,21 @@ docstyle:	## Check the docstring style using Docstyle checker
 ruff:	## Check source code using Ruff linter
 	uv run ruff check src tests --per-file-ignores=tests/*:S101 --per-file-ignores=scripts/*:S101
 
+lint-openapi: ## Lint docs/openapi.json (Spectral OAS ruleset; fail on error)
+	@if command -v npx >/dev/null 2>&1; then \
+		npx --yes @stoplight/spectral-cli@6 lint docs/openapi.json --fail-severity error --display-only-failures; \
+	else \
+		echo "lint-openapi: skipping Spectral (npx not found). Install Node.js for OpenAPI lint locally; CI still runs it."; \
+	fi
+
 verify:	## Run all linters
 	$(MAKE) black
 	$(MAKE) pylint
 	$(MAKE) pyright
 	$(MAKE) ruff
 	$(MAKE) docstyle
 	$(MAKE) check-types
+	$(MAKE) lint-openapi
 
 distribution-archives:	## Generate distribution archives to be uploaded into Python registry
 	rm -rf dist

docs/contributing/openapi-tags-and-spectral.md

@@ -0,0 +1,27 @@
+# OpenAPI tags and Spectral
+
+## Global tag list (`_OPENAPI_TAGS`)
+
+In `src/app/endpoints/`, route tags come from **`APIRouter(tags=[...])`**, which FastAPI uses when it builds the OpenAPI description for each operation.
+
+The OpenAPI document must list those tags at the top level for tools like [Spectral](https://stoplight.io/open-api/) rule **`operation-tag-defined`** to pass, so we keep **`_OPENAPI_TAGS`** in **`src/app/main.py`** and pass it into the **`FastAPI`** app as **`openapi_tags`**.
+
+**When you add a new router or change `tags=[...]` to use a new tag name**, add a matching entry to **`_OPENAPI_TAGS`** (same `name` string, plus a short `description` for the docs).
+
+The schema generator **`scripts/generate_openapi_schema.py`** passes **`tags=app.openapi_tags`** into **`get_openapi()`** so **`docs/openapi.json`** includes the top-level `tags` array. Regenerate after tag changes:
+
+```bash
+uv run scripts/generate_openapi_schema.py docs/openapi.json
+```
+
+## Linting (`make lint-openapi`)
+
+Spectral is configured in **`.spectral.yaml`** (extends `spectral:oas`). Run:
+
+```bash
+make lint-openapi
+```
+
+This is part of **`make verify`**. If **`npx`** is not on your **`PATH`**, the Makefile **skips** Spectral and prints a short message so **`make verify`** can still pass; install Node.js to run the check locally. **CI** (`.github/workflows/openapi_spectral.yaml`) always runs Spectral. Failures are driven by **error**-severity rules.
+
+The rule **`oas3-valid-media-example`** (examples must match schemas) is **turned off** in **`.spectral.yaml`** because the generated spec carries many partial examples and produced hundreds of noisy warnings. Turn it back on when examples are brought in line with schemas.

docs/openapi.json

@@ -17,7 +17,7 @@
     },
     "servers": [
         {
-            "url": "http://localhost:8080/",
+            "url": "http://localhost:8080",
             "description": "Locally running service"
         }
     ],
@@ -9262,8 +9262,7 @@
                                 "schema": {
                                     "type": "string"
                                 },
-                                "example": "event: response.created\ndata: {\"type\":\"response.created\",\"sequence_number\":0,\"response\":{\"id\":\"resp_abc\",\"object\":\"response\",\"created_at\":1704067200,\"status\":\"in_progress\",\"model\":\"openai/gpt-4o-mini\",\"output\":[],\"store\":true,\"text\":{\"format\":{\"type\":\"text\"}},\"conversation\":\"0d21ba731f21f798dc9680125d5d6f49\",\"available_quotas\":{},\"output_text\":\"\"}}\n\nevent: response.output_item.added\ndata: {\"type\":\"response.output_item.added\",\"sequence_number\":1,\"response_id\":\"resp_abc\",\"output_index\":0,\"item\":{\"id\":\"msg_abc\",\"type\":\"message\",\"status\":\"in_progress\",\"role\":\"assistant\",\"content\":[]}}\n\n...\n\nevent: response.completed\ndata: {\"type\":\"response.completed\",\"sequence_number\":30,\"response\":{\"id\":\"resp_abc\",\"object\":\"response\",\"created_at\":1704067200,\"status\":\"completed\",\"model\":\"openai/gpt-4o-mini\",\"output\":[{\"id\":\"msg_abc\",\"type\":\"message\",\"status\":\"completed\",\"role\":\"assistant\",\"content\":[{\"type\":\"output_text\",\"text\":\"Hello! How can I help?\",\"annotations\":[]}]}],\"store\":true,\"text\":{\"format\":{\"type\":\"text\"}},\"usage\":{\"input_tokens\":10,\"output_tokens\":6,\"total_tokens\":16,\"input_tokens_details\":{\"cached_tokens\":0},\"output_tokens_details\":{\"reasoning_tokens\":0}},\"conversation\":\"0d21ba731f21f798dc9680125d5d6f49\",\"available_quotas\":{\"daily\":1000,\"monthly\":50000},\"output_text\":\"Hello! How can I help?\"}}\n\ndata: [DONE]\n\n",
-                                "description": "SSE stream of events"
+                                "example": "event: response.created\ndata: {\"type\":\"response.created\",\"sequence_number\":0,\"response\":{\"id\":\"resp_abc\",\"object\":\"response\",\"created_at\":1704067200,\"status\":\"in_progress\",\"model\":\"openai/gpt-4o-mini\",\"output\":[],\"store\":true,\"text\":{\"format\":{\"type\":\"text\"}},\"conversation\":\"0d21ba731f21f798dc9680125d5d6f49\",\"available_quotas\":{},\"output_text\":\"\"}}\n\nevent: response.output_item.added\ndata: {\"type\":\"response.output_item.added\",\"sequence_number\":1,\"response_id\":\"resp_abc\",\"output_index\":0,\"item\":{\"id\":\"msg_abc\",\"type\":\"message\",\"status\":\"in_progress\",\"role\":\"assistant\",\"content\":[]}}\n\n...\n\nevent: response.completed\ndata: {\"type\":\"response.completed\",\"sequence_number\":30,\"response\":{\"id\":\"resp_abc\",\"object\":\"response\",\"created_at\":1704067200,\"status\":\"completed\",\"model\":\"openai/gpt-4o-mini\",\"output\":[{\"id\":\"msg_abc\",\"type\":\"message\",\"status\":\"completed\",\"role\":\"assistant\",\"content\":[{\"type\":\"output_text\",\"text\":\"Hello! How can I help?\",\"annotations\":[]}]}],\"store\":true,\"text\":{\"format\":{\"type\":\"text\"}},\"usage\":{\"input_tokens\":10,\"output_tokens\":6,\"total_tokens\":16,\"input_tokens_details\":{\"cached_tokens\":0},\"output_tokens_details\":{\"reasoning_tokens\":0}},\"conversation\":\"0d21ba731f21f798dc9680125d5d6f49\",\"available_quotas\":{\"daily\":1000,\"monthly\":50000},\"output_text\":\"Hello! How can I help?\"}}\n\ndata: [DONE]\n\n"
                             }
                         }
                     },
@@ -10600,15 +10599,31 @@
                 "tags": [
                     "a2a"
                 ],
-                "summary": "Handle A2A Jsonrpc",
-                "description": "Handle A2A JSON-RPC requests following the A2A protocol specification.\n\nThis endpoint uses the DefaultRequestHandler from the A2A SDK to handle\nall JSON-RPC requests including message/send, message/stream, etc.\n\nThe A2A SDK application is created per-request to include authentication\ncontext while still leveraging FastAPI's authorization middleware.\n\nAutomatically detects streaming requests (message/stream JSON-RPC method)\nand returns a StreamingResponse to enable real-time chunk delivery.\n\nArgs:\n    request: FastAPI request object\n    auth: Authentication tuple\n    mcp_headers: MCP headers for context propagation\n\nReturns:\n    JSON-RPC response or streaming response",
+                "summary": "Handle A2A JSON-RPC GET",
+                "description": "Handle GET on /a2a for A2A JSON-RPC requests following the A2A protocol specification.",
                 "operationId": "handle_a2a_jsonrpc_a2a_get",
                 "responses": {
                     "200": {
-                        "description": "Successful Response",
+                        "description": "Successful response",
                         "content": {
                             "application/json": {
-                                "schema": {}
+                                "schema": {
+                                    "type": "object",
+                                    "description": "JSON-RPC 2.0 response or A2A-over-HTTP payload"
+                                },
+                                "example": {
+                                    "jsonrpc": "2.0",
+                                    "id": "1",
+                                    "result": {}
+                                }
+                            },
+                            "text/event-stream": {
+                                "schema": {
+                                    "type": "string",
+                                    "format": "text/event-stream",
+                                    "description": "Server-Sent Events stream when the JSON-RPC method is message/stream"
+                                },
+                                "example": "data: {\"jsonrpc\":\"2.0\",\"id\":\"1\",\"result\":{}}\n\n"
                             }
                         }
                     }
@@ -10618,15 +10633,31 @@
                 "tags": [
                     "a2a"
                 ],
-                "summary": "Handle A2A Jsonrpc",
-                "description": "Handle A2A JSON-RPC requests following the A2A protocol specification.\n\nThis endpoint uses the DefaultRequestHandler from the A2A SDK to handle\nall JSON-RPC requests including message/send, message/stream, etc.\n\nThe A2A SDK application is created per-request to include authentication\ncontext while still leveraging FastAPI's authorization middleware.\n\nAutomatically detects streaming requests (message/stream JSON-RPC method)\nand returns a StreamingResponse to enable real-time chunk delivery.\n\nArgs:\n    request: FastAPI request object\n    auth: Authentication tuple\n    mcp_headers: MCP headers for context propagation\n\nReturns:\n    JSON-RPC response or streaming response",
-                "operationId": "handle_a2a_jsonrpc_a2a_get",
+                "summary": "Handle A2A JSON-RPC POST",
+                "description": "Handle POST on /a2a for A2A JSON-RPC requests following the A2A protocol specification.",
+                "operationId": "handle_a2a_jsonrpc_a2a_post",
                 "responses": {
                     "200": {
-                        "description": "Successful Response",
+                        "description": "Successful response",
                         "content": {
                             "application/json": {
-                                "schema": {}
+                                "schema": {
+                                    "type": "object",
+                                    "description": "JSON-RPC 2.0 response or A2A-over-HTTP payload"
+                                },
+                                "example": {
+                                    "jsonrpc": "2.0",
+                                    "id": "1",
+                                    "result": {}
+                                }
+                            },
+                            "text/event-stream": {
+                                "schema": {
+                                    "type": "string",
+                                    "format": "text/event-stream",
+                                    "description": "Server-Sent Events stream when the JSON-RPC method is message/stream"
+                                },
+                                "example": "data: {\"jsonrpc\":\"2.0\",\"id\":\"1\",\"result\":{}}\n\n"
                             }
                         }
                     }
@@ -20478,5 +20509,103 @@
                 ]
             }
         }
-    }
+    },
+    "tags": [
+        {
+            "name": "a2a",
+            "description": "Agent-to-Agent (A2A) protocol."
+        },
+        {
+            "name": "authorized",
+            "description": "Authorization probe."
+        },
+        {
+            "name": "config",
+            "description": "Service configuration."
+        },
+        {
+            "name": "conversations_v1",
+            "description": "Conversations API v1."
+        },
+        {
+            "name": "conversations_v2",
+            "description": "Conversations API v2."
+        },
+        {
+            "name": "feedback",
+            "description": "User feedback."
+        },
+        {
+            "name": "health",
+            "description": "Health and readiness probes."
+        },
+        {
+            "name": "info",
+            "description": "Service information."
+        },
+        {
+            "name": "mcp-auth",
+            "description": "MCP client authentication options."
+        },
+        {
+            "name": "mcp-servers",
+            "description": "MCP server registration."
+        },
+        {
+            "name": "metrics",
+            "description": "Prometheus metrics."
+        },
+        {
+            "name": "models",
+            "description": "LLM models."
+        },
+        {
+            "name": "prompts",
+            "description": "Prompt management."
+        },
+        {
+            "name": "providers",
+            "description": "Inference providers."
+        },
+        {
+            "name": "query",
+            "description": "Non-streaming query."
+        },
+        {
+            "name": "rags",
+            "description": "RAG configuration."
+        },
+        {
+            "name": "responses",
+            "description": "OpenAI-compatible Responses API."
+        },
+        {
+            "name": "rlsapi-v1",
+            "description": "RLS API v1 (inference)."
+        },
+        {
+            "name": "root",
+            "description": "Service root."
+        },
+        {
+            "name": "shields",
+            "description": "Safety shields."
+        },
+        {
+            "name": "streaming_query",
+            "description": "Streaming query (SSE)."
+        },
+        {
+            "name": "streaming_query_interrupt",
+            "description": "Streaming interrupt."
+        },
+        {
+            "name": "tools",
+            "description": "Tools."
+        },
+        {
+            "name": "vector-stores",
+            "description": "Vector stores and files."
+        }
+    ]
 }

scripts/generate_openapi_schema.py

@@ -85,6 +85,7 @@ def read_version_from_pyproject():
         license_info=app.license_info,
         servers=app.servers,
         contact=app.contact,
+        tags=app.openapi_tags,
     )
 
     # dump the schema into file