minor fixes

app.py

@@ -7,6 +7,9 @@ import json
 from typing import Dict, Any, Optional, List
 import logging
 
+from fastapi import Body, FastAPI
+from pydantic import BaseModel, ConfigDict
+
 # Add current directory to Python path
 sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
 
@@ -477,5 +480,80 @@ with gr.Blocks(title="CodeGuardian Environment - RL Enhanced") as demo:
         outputs=[reset_btn, baseline_btn, episode_btn, perf_btn, plot_btn, action_group, detect_group, risk_group, fix_group]
     )
 
+# OpenEnv / TRL-style HTTP API (POST /reset, etc.) for automated Space checks.
+# Gradio alone does not expose these routes and returns 405 Method Not Allowed.
+
+
+class OpenEnvResetRequest(BaseModel):
+    model_config = ConfigDict(extra="allow")
+
+    seed: Optional[int] = None
+    episode_id: Optional[str] = None
+
+
+class OpenEnvResetResponse(BaseModel):
+    model_config = ConfigDict(extra="forbid")
+
+    observation: Dict[str, Any]
+    reward: float = 0
+    done: bool = False
+
+
+class OpenEnvStepRequest(BaseModel):
+    model_config = ConfigDict(extra="allow")
+
+    action: Dict[str, Any]
+    timeout_s: Optional[float] = None
+    request_id: Optional[str] = None
+
+
+class OpenEnvStepResponse(BaseModel):
+    model_config = ConfigDict(extra="forbid")
+
+    observation: Dict[str, Any]
+    reward: Optional[float] = None
+    done: bool = False
+
+
+app = FastAPI(title="CodeGuardian OpenEnv")
+
+
+@app.post("/reset", response_model=OpenEnvResetResponse)
+async def openenv_reset(
+    _body: OpenEnvResetRequest = Body(default_factory=OpenEnvResetRequest),
+) -> OpenEnvResetResponse:
+    obs = space.env.reset()
+    space.current_obs = obs
+    return OpenEnvResetResponse(
+        observation=obs.model_dump(),
+        reward=0,
+        done=False,
+    )
+
+
+@app.post("/step", response_model=OpenEnvStepResponse)
+async def openenv_step(request: OpenEnvStepRequest) -> OpenEnvStepResponse:
+    action = Action.model_validate(request.action)
+    next_obs, reward_obj, done = space.env.step(action)
+    space.current_obs = next_obs
+    obs_dict = next_obs.model_dump() if next_obs is not None else {}
+    return OpenEnvStepResponse(
+        observation=obs_dict,
+        reward=float(reward_obj.value),
+        done=done,
+    )
+
+
+@app.get("/state")
+async def openenv_state() -> Dict[str, Any]:
+    return space.env.state()
+
+
+demo.queue()
+app = gr.mount_gradio_app(app, demo, path="/")
+
 if __name__ == "__main__":
-    demo.launch(server_name="0.0.0.0", server_port=7860)
+    import uvicorn
+
+    PORT = int(os.environ.get("PORT", "7860"))
+    uvicorn.run("app:app", host="0.0.0.0", port=PORT)

requirements.txt

@@ -4,5 +4,7 @@ gymnasium>=0.28.0
 numpy>=1.24.0
 python-dotenv>=1.0.0
 flask>=3.0.0
+fastapi>=0.110.0
+uvicorn[standard]>=0.27.0
 gradio>=4.0.0
 plotly