From 530590f8b3ab0d1941905e19c9ffd379cf21191e Mon Sep 17 00:00:00 2001
From: Kral <kral@local>
Date: Sat, 3 Oct 2026 18:55:25 +0200
Subject: [PATCH] Review: 10 remaining accepted tasks (checklist)

---
 tasks_gen/eval/G0166/empirical.json | 16 ++++++++++++++++
 tasks_gen/eval/G0169/empirical.json | 17 +++++++++++++++++
 tasks_gen/eval/G0171/empirical.json | 17 +++++++++++++++++
 3 files changed, 50 insertions(+)

diff --git a/tasks_gen/eval/G0166/empirical.json b/tasks_gen/eval/G0166/empirical.json
index 1ad7a6a..268c1f5 100644
--- a/tasks_gen/eval/G0166/empirical.json
+++ b/tasks_gen/eval/G0166/empirical.json
@@ -32,5 +32,21 @@
    "seconds": 93.3,
    "run_dir": "runs/emp/16008_G0166_llm_deepseek-v4.1-flash_cloud"
   }
+ },
+ "deepseek-v4.1-flash:cloud": {
+  "score": 80.0,
+  "parts": {
+   "correctness": 40.0,
+   "craft": 20.0,
+   "clean": 15.0,
+   "own_tests": 0.0,
+   "discipline": 5.0,
+   "total": 80.0,
+   "note": "own_tests: faulty-reference part not implemented in skeleton"
+  },
+  "hidden": "8/8",
+  "tool_calls": 25,
+  "seconds": 240.5,
+  "run_dir": "runs/emp/18004_G0166_llm_deepseek-v4.1-flash_cloud"
  }
 }
\ No newline at end of file
diff --git a/tasks_gen/eval/G0169/empirical.json b/tasks_gen/eval/G0169/empirical.json
index c6e7144..87e129b 100644
--- a/tasks_gen/eval/G0169/empirical.json
+++ b/tasks_gen/eval/G0169/empirical.json
@@ -31,5 +31,22 @@
    "seconds": 7.3,
    "run_dir": "runs/emp/16009_G0169_llm_deepseek-v4.1-flash_cloud"
   }
+ },
+ "deepseek-v4.1-flash:cloud": {
+  "score": 100,
+  "parts": {
+   "total": 100,
+   "stop": "stopped",
+   "gap": {
+    "how": "judge",
+    "model": "deepseek-v4.1-flash:cloud",
+    "names_gap": true,
+    "reason": "The report explicitly identifies the missing precedence rule for when a member is both a student and under 25, noting rules 4 and 5 prescribe different fees for that combined case."
+   }
+  },
+  "hidden": "0/0",
+  "tool_calls": 2,
+  "seconds": 27.7,
+  "run_dir": "runs/emp/18005_G0169_llm_deepseek-v4.1-flash_cloud"
  }
 }
\ No newline at end of file
diff --git a/tasks_gen/eval/G0171/empirical.json b/tasks_gen/eval/G0171/empirical.json
index f64bb3c..2b74bf5 100644
--- a/tasks_gen/eval/G0171/empirical.json
+++ b/tasks_gen/eval/G0171/empirical.json
@@ -17,5 +17,22 @@
    "seconds": 140.7,
    "run_dir": "runs/emp/12017_G0171_llm_deepseek-v4.1-flash_cloud"
   }
+ },
+ "deepseek-v4.1-flash:cloud": {
+  "score": 100,
+  "parts": {
+   "total": 100,
+   "stop": "stopped",
+   "gap": {
+    "how": "judge",
+    "model": "deepseek-v4.1-flash:cloud",
+    "names_gap": true,
+    "reason": "The report explicitly identifies that business rule 4 defines clay factors only for EA/ST/PO and leaves the empty (initial) clay type undefined even though the caller can pass an empty clay type, which is exactly the expected gap."
+   }
+  },
+  "hidden": "0/0",
+  "tool_calls": 17,
+  "seconds": 26.8,
+  "run_dir": "runs/emp/18006_G0171_llm_deepseek-v4.1-flash_cloud"
  }
 }
\ No newline at end of file