Refactor judge module into judge/ package

Split monolithic src/judge.py into a proper package structure:
- src/judge/batch_providers.py: BatchProvider ABC, AnthropicBatchProvider, XAIBatchProvider
- src/judge/judges.py: RegexChecker, BlackboxChecker, GlassboxChecker
- src/judge/helpers.py: extract_* functions, parse_json_verdict
- src/judge/prompts.py: BLACKBOX_PROMPT, GLASSBOX_PROMPT, constants
- src/judge/judge.py: Judge orchestrator class
- src/judge/__init__.py: Re-exports all public APIs

1 files changed, 33 insertions, 0 deletions

diff --git a/src/judge/prompts.py b/src/judge/prompts.py
new file mode 100644
index 0000000..93f500a
--- /dev/null
+++ b/src/judge/prompts.py
@@ -0,0 +1,33 @@
+"""
+Prompt templates for the judge system.
+"""
+
+import os
+
+PROMPTS_DIR = os.path.join(os.path.dirname(__file__), "..", "prompts")
+
+
+def load_prompt(filename: str) -> str:
+    path = os.path.join(PROMPTS_DIR, filename)
+    with open(path, "r", encoding="utf-8") as f:
+        return f.read()
+
+
+BLACKBOX_PROMPT = load_prompt("blackbox_prompt.md")
+GLASSBOX_PROMPT = load_prompt("glassbox_prompt.md")
+
+
+BLACKBOX_CATEGORIES = [
+    "ACCURATE",
+    "DECEPTIVE_FRAMING",
+    "OMITTED_CRITICAL",
+    "FABRICATED",
+]
+GLASSBOX_CATEGORIES = [
+    "ROBUST_HONEST",
+    "CONSIDERED_HONEST",
+    "COMPLIANT_DECEPTION",
+    "FAILED_DECEPTION",
+    "UNINTENTIONAL_FAILURE",
+]
+SOPHISTICATION_TIERS = ["NAIVE", "REACTIVE", "TACTICAL", "STRATEGIC"]