ci: update evalbench pipeline trigger and add missing scorers

.github/labels.yaml

@@ -84,3 +84,7 @@
 - name: 'release-please:force-run'
   color: bdca82
   description: Manually trigger the release please workflow on a PR.
+
+- name: 'ci:run-evals'
+  color: 4285f4
+  description: Manually trigger the evaluation CI pipeline on a PR.

cloudbuild.yaml

@@ -14,6 +14,8 @@
 
 options:
   logging: CLOUD_LOGGING_ONLY
+  pool:
+    name: 'projects/ext-test-alloydb/locations/us-central1/workerPools/alloydb-evals-pool'
 
 steps:
 
@@ -27,12 +29,7 @@ steps:
       - |
         set -e
 
-        # Only run on release branches
-        if [[ "$_HEAD_BRANCH" != release-please-* ]]; then
-          echo "Not a release-please branch. Exiting."
-          exit 0
-        fi
-        echo "Release branch detected. Fetching PR data from GitHub API..."
+        echo "Fetching PR data from GitHub API..."
 
         # Fetch PR data and status code
         HTTP_STATUS=$(curl -s -o pr_data.json -w "%{http_code}" -H "Authorization: token $$GITHUB_TOKEN" \
@@ -46,29 +43,38 @@ steps:
 
         PR_DATA=$(cat pr_data.json)
 
-        # Extract labels and title from PR data (Use $$ to escape bash variables)
-        PR_LABELS=$(echo "$$PR_DATA" | jq -r '[.labels[].name] | join(",")')
+        # Extract title from PR data (Use $$ to escape bash variables)
         PR_TITLE=$(echo "$$PR_DATA" | jq -r '.title')
 
-        # Determine Release Version (Use double quotes and $$ for bash variables)
-        if [[ "$$PR_LABELS" == *"autorelease: pending"* ]]; then
+        # Check if execution labels are present using exact matching via jq
+        if ! jq -e '.labels | any(.name == "autorelease: pending" or .name == "ci:run-evals")' pr_data.json > /dev/null; then
+          echo "PR does not have 'autorelease: pending' or 'ci:run-evals' label. Skipping execution."
+          exit 0
+        fi
+        echo "Execution label detected. Processing release version context..."
+
+        # Determine Release Version based on branch name
+        if [[ "$_HEAD_BRANCH" == release-please-* ]]; then
           if [[ "$$PR_TITLE" =~ release\ ([0-9]+\.[0-9]+\.[0-9]+) ]]; then
             export RELEASE_VERSION="$${BASH_REMATCH[1]}"
           else
-            export RELEASE_VERSION="unknown"
+            export RELEASE_VERSION="pr-$_PR_NUMBER-release-unknown"
           fi
         else
-          export RELEASE_VERSION="unknown"
+          export RELEASE_VERSION="pr-$_PR_NUMBER-ci-run-evals"
         fi
 
         # Workaround for evalbench bug: settings are only applied if path basename matches extension ID
         ln -s /workspace /workspace/alloydb
         cd /evalbench
 
+        # Evalbench environment variables
         export EVAL_GCP_PROJECT_ID=$PROJECT_ID
         export EVAL_GCP_PROJECT_REGION=$_ALLOYDB_REGION
         export EVAL_REPORTING_PROJECT=$_EVAL_REPORTING_PROJECT
-
+        export GOOGLE_CLOUD_PROJECT=$PROJECT_ID
+
+        # AlloyDB PostgreSQL specific environment variables
         export ALLOYDB_POSTGRES_PROJECT=$PROJECT_ID
         export ALLOYDB_POSTGRES_REGION=$_ALLOYDB_REGION
         export ALLOYDB_POSTGRES_CLUSTER=$_ALLOYDB_CLUSTER

evals/dataset.json

@@ -2,7 +2,7 @@
   "scenarios": [
     {
       "id": "alloydb-debug-instance",
-      "starting_prompt": "Check my AlloyDB databases in project ${EVAL_GCP_PROJECT_ID}.",
+      "starting_prompt": "Check my AlloyDB clusters and instances in project ${EVAL_GCP_PROJECT_ID} to verify their availability.",
       "conversation_plan": "Ask the agent to list all AlloyDB instances in the project. Once all instances are listed, if ${ALLOYDB_POSTGRES_INSTANCE} exists, get its details and validate it is RUNNABLE.",
       "expected_trajectory": [
         "list_instances",

evals/model_config.yaml

@@ -17,8 +17,11 @@ generator: gemini_cli
 env:
   GOOGLE_CLOUD_PROJECT: "${GOOGLE_CLOUD_PROJECT}"
   GOOGLE_CLOUD_LOCATION: "global"
+  GOOGLE_CLOUD_REGION: "${EVAL_GCP_PROJECT_REGION}"
+  CLOUDSDK_COMPUTE_REGION: "${EVAL_GCP_PROJECT_REGION}"
   GOOGLE_GENAI_USE_VERTEXAI: "true"
   GEMINI_CLI_TRUST_WORKSPACE: "true"
+
 setup:
   extensions:
     "/workspace/alloydb":

evals/run_config.yaml

@@ -25,12 +25,16 @@ scorers:
     model_config: /workspace/evals/gemini_2.5_pro_model.yaml
   behavioral_metrics:
     model_config: /workspace/evals/gemini_2.5_pro_model.yaml
+  skills_best_practices:
+    model_config: /workspace/evals/gemini_2.5_pro_model.yaml
+    skills_dir: /workspace/alloydb/skills
 
   # Performance
   turn_count: {}
   end_to_end_latency: {}
   tool_call_latency: {}
   token_consumption: {}
+  skills_trajectory: {}
 
 reporting:
   bigquery: