diff --git a/ci-operator/config/openshift-eng/ai-helpers/openshift-eng-ai-helpers-main.yaml b/ci-operator/config/openshift-eng/ai-helpers/openshift-eng-ai-helpers-main.yaml
index 54500910ac182..e7d2701937577 100644
--- a/ci-operator/config/openshift-eng/ai-helpers/openshift-eng-ai-helpers-main.yaml
+++ b/ci-operator/config/openshift-eng/ai-helpers/openshift-eng-ai-helpers-main.yaml
@@ -67,6 +67,8 @@ tests:
     env:
       EVAL_CASES: case-003
       EVAL_CONFIG: plugins/ci/evals/eval-payload-analysis.yaml
+      EVAL_HARNESS_REF: eval-stop-guard-hook
+      EVAL_HARNESS_REPO: https://github.com/stbenjam/agent-eval-harness.git
       EVAL_MAX_TURNS: "250"
       EVAL_MODEL: claude-opus-4-6
       EVAL_SETUP_SCRIPT: plugins/ci/evals/scripts/extract-payload-analysis-snapshots.sh
diff --git a/ci-operator/step-registry/openshift/claude/agent-eval/openshift-claude-agent-eval-commands.sh b/ci-operator/step-registry/openshift/claude/agent-eval/openshift-claude-agent-eval-commands.sh
index 0851c71909fad..e80bf104e0449 100755
--- a/ci-operator/step-registry/openshift/claude/agent-eval/openshift-claude-agent-eval-commands.sh
+++ b/ci-operator/step-registry/openshift/claude/agent-eval/openshift-claude-agent-eval-commands.sh
@@ -109,8 +109,11 @@ fi
 echo ""
 echo "=== Installing plugins ==="
 EVAL_HARNESS_DIR="/tmp/agent-eval-harness"
-git clone --depth 1 https://github.com/opendatahub-io/agent-eval-harness.git "${EVAL_HARNESS_DIR}"
-echo "agent-eval-harness cloned."
+# EVAL_HARNESS_REPO / EVAL_HARNESS_REF let a PR rehearse the eval against a
+# not-yet-merged branch of the harness. They default to the released main.
+EVAL_HARNESS_REPO="${EVAL_HARNESS_REPO:-https://github.com/opendatahub-io/agent-eval-harness.git}"
+git clone --depth 1 ${EVAL_HARNESS_REF:+--branch "${EVAL_HARNESS_REF}"} "${EVAL_HARNESS_REPO}" "${EVAL_HARNESS_DIR}"
+echo "agent-eval-harness cloned from ${EVAL_HARNESS_REPO} (ref: ${EVAL_HARNESS_REF:-default})."
 
 # -----------------------------------------------------------------------
 # Run optional setup script (e.g. extract snapshots, populate fixtures)
diff --git a/ci-operator/step-registry/openshift/claude/agent-eval/openshift-claude-agent-eval-ref.yaml b/ci-operator/step-registry/openshift/claude/agent-eval/openshift-claude-agent-eval-ref.yaml
index 7868e87b9d868..43c4c7cdd43e4 100644
--- a/ci-operator/step-registry/openshift/claude/agent-eval/openshift-claude-agent-eval-ref.yaml
+++ b/ci-operator/step-registry/openshift/claude/agent-eval/openshift-claude-agent-eval-ref.yaml
@@ -10,6 +10,10 @@ ref:
     name: claude-payload-agent-github-token
     mount_path: /var/run/github-token
   env:
+  - name: EVAL_HARNESS_REPO
+    default: "https://github.com/opendatahub-io/agent-eval-harness.git"
+  - name: EVAL_HARNESS_REF
+    default: ""
   - name: EVAL_CONFIG
     default: "eval.yaml"
   - name: EVAL_MODEL