From 3ff2e2ce11f34aec234e151eedcca9a8dcc4e976 Mon Sep 17 00:00:00 2001
From: steven_li <liyuxuan2001@hotmail.com>
Date: Fri, 22 May 2026 11:28:19 +0800
Subject: [PATCH] fix(task): complete evidence rendering contract

---
 app-instance/backend/beaver/tasks/evidence.py |  3 ++-
 .../backend/tests/unit/test_task_evidence.py  | 22 ++++++++++++++-----
 2 files changed, 18 insertions(+), 7 deletions(-)

diff --git a/app-instance/backend/beaver/tasks/evidence.py b/app-instance/backend/beaver/tasks/evidence.py
index 9717041..02ccb20 100644
--- a/app-instance/backend/beaver/tasks/evidence.py
+++ b/app-instance/backend/beaver/tasks/evidence.py
@@ -79,7 +79,6 @@ class EvidenceBuilder:
 
     def build_run_evidence(
         self,
-        *,
         session_id: str,
         run_id: str,
         output_text: str,
@@ -175,6 +174,8 @@ def _render_tool_evidence(item: ToolEvidence) -> str:
         metadata.append(f"url={item.url}")
     if item.title:
         metadata.append(f"title={item.title}")
+    if item.created_at:
+        metadata.append(f"created_at={item.created_at}")
     return "\n".join([header, *metadata, item.content])
 
 
diff --git a/app-instance/backend/tests/unit/test_task_evidence.py b/app-instance/backend/tests/unit/test_task_evidence.py
index 4ba494c..6206642 100644
--- a/app-instance/backend/tests/unit/test_task_evidence.py
+++ b/app-instance/backend/tests/unit/test_task_evidence.py
@@ -3,7 +3,7 @@ from __future__ import annotations
 from pathlib import Path
 
 from beaver.engine.session.manager import SessionManager
-from beaver.tasks.evidence import EvidenceBuilder, RunEvidence, TaskEvidencePacket, render_task_evidence
+from beaver.tasks.evidence import EvidenceBuilder, RunEvidence, TaskEvidencePacket, ToolEvidence, render_task_evidence
 
 
 def test_evidence_builder_preserves_full_tool_result(tmp_path: Path) -> None:
@@ -35,10 +35,10 @@ def test_evidence_builder_preserves_full_tool_result(tmp_path: Path) -> None:
     )
 
     evidence = EvidenceBuilder(session_manager).build_run_evidence(
-        session_id=session_id,
-        run_id=run_id,
-        output_text="Manchester United won 3-2.",
-        finish_reason="stop",
+        session_id,
+        run_id,
+        "Manchester United won 3-2.",
+        "stop",
     )
     rendered = render_task_evidence(
         TaskEvidencePacket(
@@ -63,7 +63,15 @@ def test_render_task_evidence_includes_failed_team_run_tool_results() -> None:
         output_text="Tool loop stopped.",
         finish_reason="max_tool_iterations",
         transcript=[],
-        tool_results=[],
+        tool_results=[
+            ToolEvidence(
+                tool_name="web_fetch",
+                tool_call_id="call-team",
+                content="Recovered partial source content.",
+                event_payload={"success": True, "created_at": "2026-05-22T12:00:00Z"},
+                created_at="2026-05-22T12:00:00Z",
+            )
+        ],
         warnings=["finish_reason=max_tool_iterations"],
     )
     packet = TaskEvidencePacket(
@@ -79,3 +87,5 @@ def test_render_task_evidence_includes_failed_team_run_tool_results() -> None:
 
     assert "finish_reason=max_tool_iterations" in rendered
     assert "partial answer" in rendered
+    assert "Recovered partial source content." in rendered
+    assert "created_at=2026-05-22T12:00:00Z" in rendered