listing cleanup

Author: ourmony

tools/listing.py

@@ -227,4 +227,4 @@ def generate_readme() -> None:
 
 
 if __name__ == "__main__":
-    generate_readme()
+    generate_readme()

tools/listing.yaml

@@ -66,6 +66,19 @@
   tasks: ["cybench"]
   tags: ["Agent"]
 
+- title: "CyberSecEval_2: A Wide-Ranging Cybersecurity Evaluation Suite for Large Language Models"
+  description: |
+    Evaluates Large Language Models for risky capabilities in cybersecurity.
+  path: src/inspect_evals/cyberseceval_2
+  arxiv: https://arxiv.org/pdf/2404.13161
+  group: Cybersecurity
+  contributors: ["its-emile"]
+  tasks: [
+    "interpreter_abuse",
+    "prompt_injection",
+    "vulnerability_exploit"
+  ]
+
 - title: "InterCode: Capture the Flag"
   description: |
     Measure expertise in coding, cryptography (i.e. binary exploitation, forensics), reverse engineering, and recognizing security vulnerabilities. Demonstrates tool use and sandboxing untrusted model code.
@@ -352,17 +365,4 @@
       "agie_sat_en",
       "agie_sat_en_without_passage",
       "agie_sat_math",
-    ]
-
-- title: "CyberSecEval_2: A Wide-Ranging Cybersecurity Evaluation Suite for Large Language Models"
-  description: |
-    Evaluates Large Language Models for risky capabilities in cybersecurity.
-  path: src/inspect_evals/cyberseceval_2
-  arxiv: https://arxiv.org/pdf/2404.13161
-  group: Cybersecurity
-  contributors: ["its-emile"]
-  tasks: [
-    "interpreter_abuse",
-    "prompt_injection",
-    "vulnerability_exploit"
-  ]
+    ]