feat: implement state interface, standardize schema IDs, and add task discovery endpoint

envs/social_stream_moderation/environment.py

@@ -66,7 +66,10 @@ class SocialStreamModerationEnv:
             step_index=self.step_index,
             total_steps=len(self.episode_posts)
         )
-        
+    def state(self) -> State:
+        """Standard OpenEnv interface to return the current observation."""
+        return self._get_state()   
+
     async def step(self, action: ModerationAction) -> Tuple[Optional[State], float, bool, Dict[str, Any]]:
         """Processes one moderation action."""
         if self.done:

openenv.yaml

@@ -1,55 +1,55 @@
-name: SocialStreamModerationEnv
-version: 1.0.0
-graders:
-  - name: basic_safety_grader
-    description: "Grader for basic safety checks"
-  - name: context_nuance_grader
-    description: "Grader for contextual and sarcastic content"
-  - name: fairness_bias_grader
-    description: "Grader for fairness and bias parity"
-
-tasks:
-  - name: "Task 1: Basic Safety"
-    difficulty: easy
-    description: "Moderate a stream of social posts with obvious violations and safe content."
-    grader: basic_safety_grader
-  - name: "Task 2: Context & Nuance"
-    difficulty: medium
-    description: "Handle sarcastic content and quotes of harmful material with condemnation."
-    grader: context_nuance_grader
-  - name: "Task 3: Fairness & Bias"
-    difficulty: hard
-    description: "Ensure fairness across user groups and adhere to stricter policy regimes."
-    grader: fairness_bias_grader
-
-schemas:
-  state:
-    type: object
-    properties:
-      post_id: { type: string }
-      text: { type: string }
-      user_history_summary: { type: string }
-      context_type: { type: string }
-      platform_policy_mode: { type: string }
-      user_group: { type: string }
-      step_index: { type: integer }
-      total_steps: { type: integer }
-  action:
-    type: string
-    enum:
-      - ALLOW
-      - ALLOW_WITH_WARNING
-      - SOFT_HIDE
-      - ESCALATE_HUMAN
-      - BAN_USER
-  info:
-    type: object
-    properties:
-      ground_truth_label: { type: string }
-      action_taken: { type: string }
-      reward: { type: number }
-      score: { type: number }
-
-paths:
-  inference: ./inference.py
-  app: ./server/app.py
+name: SocialStreamModerationEnv
+version: 1.0.0
+graders:
+  - id: basic_safety_grader
+    description: "Grader for basic safety checks"
+  - id: context_nuance_grader
+    description: "Grader for contextual and sarcastic content"
+  - id: fairness_bias_grader
+    description: "Grader for fairness and bias parity"
+
+tasks:
+  - id: "Task 1: Basic Safety"
+    difficulty: easy
+    description: "Moderate a stream of social posts with obvious violations and safe content."
+    grader_id: basic_safety_grader
+  - id: "Task 2: Context & Nuance"
+    difficulty: medium
+    description: "Handle sarcastic content and quotes of harmful material with condemnation."
+    grader_id: context_nuance_grader
+  - id: "Task 3: Fairness & Bias"
+    difficulty: hard
+    description: "Ensure fairness across user groups and adhere to stricter policy regimes."
+    grader_id: fairness_bias_grader
+
+schemas:
+  state:
+    type: object
+    properties:
+      post_id: { type: string }
+      text: { type: string }
+      user_history_summary: { type: string }
+      context_type: { type: string }
+      platform_policy_mode: { type: string }
+      user_group: { type: string }
+      step_index: { type: integer }
+      total_steps: { type: integer }
+  action:
+    type: string
+    enum:
+      - ALLOW
+      - ALLOW_WITH_WARNING
+      - SOFT_HIDE
+      - ESCALATE_HUMAN
+      - BAN_USER
+  info:
+    type: object
+    properties:
+      ground_truth_label: { type: string }
+      action_taken: { type: string }
+      reward: { type: number }
+      score: { type: number }
+
+paths:
+  inference: ./inference.py
+  app: ./server/app.py

server/app.py

@@ -767,13 +767,37 @@ async def reset_env(req: ResetRequest = Body(default=ResetRequest())):
     except ValueError as e:
         raise HTTPException(status_code=400, detail=str(e))
 
+@app.get("/tasks", tags=["🤖 Automated Benchmarking"])
+async def list_tasks():
+    """Returns the list of tasks available in the environment for discovery."""
+    return [
+        {
+            "id": "Task 1: Basic Safety",
+            "difficulty": "easy",
+            "description": "Moderate a stream of social posts with obvious violations and safe content.",
+            "grader_id": "basic_safety_grader"
+        },
+        {
+            "id": "Task 2: Context & Nuance",
+            "difficulty": "medium",
+            "description": "Handle sarcastic content and quotes of harmful material with condemnation.",
+            "grader_id": "context_nuance_grader"
+        },
+        {
+            "id": "Task 3: Fairness & Bias",
+            "difficulty": "hard",
+            "description": "Ensure fairness across user groups and adhere to stricter policy regimes.",
+            "grader_id": "fairness_bias_grader"
+        }
+    ]
+
 @app.get("/graders", tags=["🛡️ Automated Benchmarking"])
 async def list_graders():
     """Returns the list of graders available in the environment for discovery."""
     return [
-        {"name": "basic_safety_grader", "description": "Grader for basic safety checks"},
-        {"name": "context_nuance_grader", "description": "Grader for contextual and sarcastic content"},
-        {"name": "fairness_bias_grader", "description": "Grader for fairness and bias parity"}
+        {"id": "basic_safety_grader", "description": "Grader for basic safety checks"},
+        {"id": "context_nuance_grader", "description": "Grader for contextual and sarcastic content"},
+        {"id": "fairness_bias_grader", "description": "Grader for fairness and bias parity"}
     ]
 
 @app.post("/evaluate", tags=["🧪 Interactive Lab"], summary="Test Model Logic (XAI Insight)")