keda: add container-based config discovery support

Add a discovery strategy and auto_conf.yaml for the keda-operator-metrics-apiserver
component so the Agent's Kubernetes Autodiscovery listener can automatically
generate an openmetrics_endpoint configuration for it.

Environment: Datadog workspace

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Co-authored-by: Vincent Whitchurch <vincent.whitchurch@datadoghq.com>

Author: vitkyrka

keda/assets/configuration/spec.yaml

@@ -2,6 +2,12 @@ name: Keda
 fleet_configurable: true
 files:
 - name: keda.yaml
+  discovery:
+    strategies:
+    - template: discovery/openmetrics_from_ports
+      overrides:
+        port_hints:
+        - 8080
   options:
   - template: init_config
     options:
@@ -13,4 +19,11 @@ files:
         openmetrics_endpoint.value.example: http://localhost:8080/metrics
         openmetrics_endpoint.description: |
           Endpoint exposing the Keda's Prometheus metrics. For more information refer to:
-          https://keda.sh/docs/2.16/integrations/prometheus/
+          https://keda.sh/docs/2.16/integrations/prometheus/
+- name: auto_conf.yaml
+  options:
+  - template: ad_identifiers
+    overrides:
+      value.example:
+      - keda-metrics-apiserver
+  - template: auto_conf/discovery

keda/datadog_checks/keda/config_models/discovery.py

@@ -0,0 +1,42 @@
+# (C) Datadog, Inc. 2026-present
+# All rights reserved
+# Licensed under a 3-clause BSD style license (see LICENSE)
+
+# This file is autogenerated.
+# To change this file you should edit assets/configuration/spec.yaml and then run the following commands:
+#     ddev -x validate config -s <INTEGRATION_NAME>
+#     ddev -x validate models -s <INTEGRATION_NAME>
+
+from __future__ import annotations
+
+from collections.abc import Iterator
+from typing import Any
+
+from datadog_checks.base.utils.discovery import Service, candidate_ports
+from datadog_checks.keda.config_models import discovery_overrides
+from datadog_checks.keda.config_models.instance import InstanceConfig
+from datadog_checks.keda.config_models.shared import SharedConfig
+
+
+def _generated_candidates(service: Service) -> Iterator[dict[str, Any]]:
+    shared = SharedConfig.model_validate({}, context={'configured_fields': frozenset()}).model_dump(
+        by_alias=True, mode='json', exclude_none=True
+    )
+    # discovery[0]: from_ports
+    for port in candidate_ports(service, [8080]):
+        ctx = {'port': port}
+        instance_data = {
+            'openmetrics_endpoint': 'http://{service.host}:{port.number}/metrics'.format(service=service, **ctx),
+        }
+        instance = InstanceConfig.model_validate(
+            instance_data, context={'configured_fields': frozenset(instance_data)}
+        ).model_dump(by_alias=True, mode='json', exclude_none=True)
+        yield {'init_config': shared, 'instances': [instance]}
+
+
+def candidates(service: Service) -> Iterator[dict[str, Any]]:
+    override = getattr(discovery_overrides, 'candidates', None)
+    if override is None:
+        yield from _generated_candidates(service)
+    else:
+        yield from override(service, default=_generated_candidates)

keda/datadog_checks/keda/config_models/discovery_overrides.py

@@ -0,0 +1,12 @@
+# (C) Datadog, Inc. 2026-present
+# All rights reserved
+# Licensed under a 3-clause BSD style license (see LICENSE)
+
+# Override the generated discovery candidates() for this integration.
+#
+# Define a candidates(service, default) function to wrap or replace the generated
+# candidate generation. `default` is the generated generator; call it to reuse
+# the spec-driven candidates, or ignore it to replace them entirely.
+#
+# def candidates(service, default):
+#     yield from default(service)

keda/datadog_checks/keda/config_models/discovery_strategies.py

@@ -0,0 +1,18 @@
+# (C) Datadog, Inc. 2026-present
+# All rights reserved
+# Licensed under a 3-clause BSD style license (see LICENSE)
+
+# Here you can define custom (local:) discovery strategies for this integration.
+#
+# Decorate a generator with @discovery_strategy (imported from
+# datadog_checks.base.utils.discovery) and reference it from the spec discovery
+# stanza as `strategy: local:<function_name>`. The function receives the
+# discovered Service plus the inputs declared in the spec and yields one context
+# (ctx) mapping per candidate, exposing the keys listed in `provides`.
+#
+# from datadog_checks.base.utils.discovery import discovery_strategy
+#
+# @discovery_strategy(provides=('svc',))
+# def from_some_config(service, config_path):
+#     ...
+#     yield {'svc': ...}

keda/datadog_checks/keda/data/auto_conf.yaml

@@ -0,0 +1,19 @@
+## @param ad_identifiers - list of strings - required
+## A list of container identifiers that are used by Autodiscovery to identify
+## which container the check should be run against. For more information, see:
+## https://docs.datadoghq.com/agent/guide/ad_identifiers/
+#
+ad_identifiers:
+  - keda-metrics-apiserver
+
+## Enables configuration discovery
+#
+discovery: {}
+
+## Unused init configuration
+#
+init_config:
+
+## Unused instance configuration
+#
+instances: []