From 3fcdc422d03687db12ec195454784f57b9712de0 Mon Sep 17 00:00:00 2001
From: fullsend-code
 <278716306+fullsend-ai-coder[bot]@users.noreply.github.com>
Date: Thu, 23 Jul 2026 04:51:36 +0000
Subject: [PATCH 1/2] test(#5493): fold PR-to-issue negative into PR-label
 positive scenario
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Remove the standalone "PR does not trigger issue-only harness" scenario
and fold its negative assertion into the "PR label dispatches PR-only
harness" scenario. The positive scenario already opens a PR, labels it,
and waits for pr-ping to complete — the negative check piggy-backs on
the same dispatch round-trip without a separate 90s settle wait.

The merged scenario now:
- Registers both pr-ping (change_proposal) and issue-only-ping
  (work_item) with the same label, verifying entity kind filtering
- Asserts pr-ping workflow completes and dummy proves execution
- Asserts pr-ping was dispatched exactly 1 time
- Asserts issue-only-ping agent did not run

This follows the enabled/disabled pattern already established by the
"Disabled harness is not dispatched while enabled one triggers" scenario.

Closes #5493

Signed-off-by: Barak Korren <bkorren@redhat.com>
---
 .../features/dispatch/dispatch.feature        | 22 +++++++++----------
 1 file changed, 10 insertions(+), 12 deletions(-)

diff --git a/e2e/behaviour/features/dispatch/dispatch.feature b/e2e/behaviour/features/dispatch/dispatch.feature
index 85dfe38f2d..16e7c93aee 100644
--- a/e2e/behaviour/features/dispatch/dispatch.feature
+++ b/e2e/behaviour/features/dispatch/dispatch.feature
@@ -32,32 +32,28 @@ Feature: Harness CEL dispatch
     And the agent will succeed to Prove execution
     And the harness "issue-ping" was dispatched exactly 1 time
 
-  Scenario: PR does not trigger issue-only harness
-    Given a custom harness "issue-ping" with:
+  Scenario: PR label dispatches PR-only harness but not issue-only harness
+    Given a custom harness "pr-ping" with:
       """
       agent: agents/triage.md
       role: triage
-      slug: fullsend-ai-issue-ping
+      slug: fullsend-ai-pr-ping
       model: opus
       image: ghcr.io/fullsend-ai/fullsend-sandbox:latest
       trigger: |
-        event.entity.kind == "work_item"
+        event.entity.kind == "change_proposal"
         && event.transition.kind == "label_changed"
-        && event.transition.label.name == "ready-for-ping"
+        && event.transition.label.name == "ready-for-pr-ping"
       """
-    When a pull request is opened
-    Then the harness "issue-ping" agent did not run
-
-  Scenario: PR label dispatches PR-only harness
-    Given a custom harness "pr-ping" with:
+    And a custom harness "issue-only-ping" with:
       """
       agent: agents/triage.md
       role: triage
-      slug: fullsend-ai-pr-ping
+      slug: fullsend-ai-issue-only-ping
       model: opus
       image: ghcr.io/fullsend-ai/fullsend-sandbox:latest
       trigger: |
-        event.entity.kind == "change_proposal"
+        event.entity.kind == "work_item"
         && event.transition.kind == "label_changed"
         && event.transition.label.name == "ready-for-pr-ping"
       """
@@ -69,6 +65,8 @@ Feature: Harness CEL dispatch
     And the pull request is labeled "ready-for-pr-ping"
     Then the harness "pr-ping" workflow completes successfully
     And the agent will succeed to Prove PR execution
+    And the harness "pr-ping" was dispatched exactly 1 time
+    And the harness "issue-only-ping" agent did not run
 
   Scenario: Disabled harness is not dispatched while enabled one triggers
     Given a custom harness "enabled-ping" with:

From 034f8468177a41a9180a2b1f18f950a67e97cb2a Mon Sep 17 00:00:00 2001
From: Barak Korren <bkorren@redhat.com>
Date: Thu, 23 Jul 2026 11:22:07 +0300
Subject: [PATCH 2/2] ci(e2e): raise behaviour job timeout to 45m
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Healthy behaviour suites run ~24–29m and were flaking against the 30m
job ceiling after the folded dispatch scenario already passed. Align the
Makefile go test -timeout with the workflow budget.

Signed-off-by: Barak Korren <bkorren@redhat.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
---
 .github/workflows/e2e.yml | 2 +-
 Makefile                  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/e2e.yml b/.github/workflows/e2e.yml
index d1d093b3b2..64518432ba 100644
--- a/.github/workflows/e2e.yml
+++ b/.github/workflows/e2e.yml
@@ -189,7 +189,7 @@ jobs:
       !cancelled() &&
       (github.event_name != 'pull_request_target' || needs.gate.outputs.authorized == 'true')
     runs-on: ubuntu-24.04
-    timeout-minutes: 30
+    timeout-minutes: 45
     permissions:
       contents: read
       id-token: write
diff --git a/Makefile b/Makefile
index ddc1a858ae..2d48a58ae0 100644
--- a/Makefile
+++ b/Makefile
@@ -171,7 +171,7 @@ e2e-test:
 	go test -tags e2e -v -count=1 -timeout 30m ./e2e/admin/
 
 behaviour-test:
-	go test -tags behaviour -v -count=1 -timeout 30m ./e2e/behaviour/
+	go test -tags behaviour -v -count=1 -timeout 45m ./e2e/behaviour/
 
 # Functional agent evals — run agents against ephemeral GitHub repos and judge results.
 # Required env: EVAL_ORG (GitHub org for ephemeral repos), plus GCP creds for Vertex AI.