Update codebase with various modifications

.agent/rules/graphify.md

@@ -0,0 +1,9 @@
+## graphify
+
+This project has a graphify knowledge graph at graphify-out/.
+
+Rules:
+- Before answering architecture or codebase questions, read graphify-out/GRAPH_REPORT.md for god nodes and community structure
+- If graphify-out/wiki/index.md exists, navigate it instead of reading raw files
+- If the graphify MCP server is active, utilize tools like `query_graph`, `get_node`, and `shortest_path` for precise architecture navigation instead of falling back to `grep`
+- After modifying code files in this session, run `python3 -c "from graphify.watch import _rebuild_code; from pathlib import Path; _rebuild_code(Path('.'))"` to keep the graph current

.agent/workflows/graphify.md

@@ -0,0 +1,8 @@
+# Workflow: graphify
+**Command:** /graphify
+**Description:** Turn any folder of files into a navigable knowledge graph
+
+## Steps
+Follow the graphify skill installed at ~/.agent/skills/graphify/SKILL.md to run the full pipeline.
+
+If no path argument is given, use `.` (current directory).

.cursor/rules/graphify.mdc

@@ -0,0 +1,10 @@
+---
+description: graphify knowledge graph context
+alwaysApply: true
+---
+
+This project has a graphify knowledge graph at graphify-out/.
+
+- Before answering architecture or codebase questions, read graphify-out/GRAPH_REPORT.md for god nodes and community structure
+- If graphify-out/wiki/index.md exists, navigate it instead of reading raw files
+- After modifying code files in this session, run `python3 -c "from graphify.watch import _rebuild_code; from pathlib import Path; _rebuild_code(Path('.'))"` to keep the graph current

.opencode/plugins/graphify.js

@@ -0,0 +1,22 @@
+// graphify OpenCode plugin
+// Injects a knowledge graph reminder before bash tool calls when the graph exists.
+import { existsSync } from "fs";
+import { join } from "path";
+
+export const GraphifyPlugin = async ({ directory }) => {
+  let reminded = false;
+
+  return {
+    "tool.execute.before": async (input, output) => {
+      if (reminded) return;
+      if (!existsSync(join(directory, "graphify-out", "graph.json"))) return;
+
+      if (input.tool === "bash") {
+        output.args.command =
+          'echo "[graphify] Knowledge graph available. Read graphify-out/GRAPH_REPORT.md for god nodes and architecture context before searching files." && ' +
+          output.args.command;
+        reminded = true;
+      }
+    },
+  };
+};

Dockerfile

@@ -19,11 +19,11 @@ ENV PYTHONPATH="/app"
 ENV PYTHONUNBUFFERED=1
 
 # Expose port
-EXPOSE 8000
+EXPOSE 7860
 
-# Health check
-HEALTHCHECK --interval=30s --timeout=3s --start-period=5s --retries=3 \
-    CMD curl -f http://localhost:8000/health || exit 1
+# Health check - generous timeouts for Hugging Face Spaces
+HEALTHCHECK --interval=30s --timeout=10s --start-period=120s --retries=10 \
+    CMD curl -f http://localhost:7860/health || exit 1
 
 # Run server (the app.py lifespan prints the full banner)
-CMD ["python", "-m", "uvicorn", "server.app:app", "--host", "0.0.0.0", "--port", "8000"]
+CMD ["python", "-m", "uvicorn", "server.app:app", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -1,3 +1,17 @@
+---
+title: Mission Control
+emoji: "☁️"
+colorFrom: blue
+colorTo: indigo
+sdk: docker
+tags:
+  - finops
+  - openenv
+  - fastapi
+  - observability
+  - dashboard
+---
+
 # 🛡️ MissionCtrl — AI Oversight Fleet Environment
 
 > *Every LLM agent fleet will hallucinate. MissionCtrl trains the overseer to catch them.*
@@ -204,7 +218,7 @@ docker build -t missionctrl .
 docker run -p 8000:8000 --name missionctrl missionctrl
 
 # 3. Run the baseline agent (in another terminal)
-docker exec -it missionctrl python inference.py
+docker exec -it missionctrl python client.py
 
 # 4. Watch the dashboard
 open http://localhost:8000/dashboard
@@ -222,8 +236,8 @@ python -m uvicorn server.app:app --host 0.0.0.0 --port 8000
 # Configure API keys
 cp .env.example .env   # fill in your LLM provider keys
 
-# Run inference
-python inference.py
+# Run inference (OpenEnv canonical entrypoint)
+python client.py
 
 # Run tests
 pytest tests/ -v
@@ -301,7 +315,8 @@ missionctrl/
 ├── openenv.yaml           # OpenEnv manifest
 ├── pyproject.toml         # Python project config
 ├── Dockerfile             # Single-container deployment
-├── inference.py           # Baseline LLM agent + cross-episode memory
+├── client.py              # OpenEnv-required baseline evaluator entrypoint
+├── inference.py           # Backward-compatible wrapper to client.main()
 ├── .env.example           # API key template
 ├── server/
 │   ├── app.py             # FastAPI server (6 endpoints + dashboard)
@@ -343,15 +358,53 @@ missionctrl/
 | Method | Endpoint | Description |
 |--------|----------|-------------|
 | `GET` | `/` | Status heartbeat |
-| `GET` | `/health` | Readiness check (`{"status": "ok"}`) |
+| `GET` | `/health` | Readiness check (`{"healthy": true, "env": "missionctrl"}`) |
 | `POST` | `/reset` | `{"task_id": "easy"}` → Reset environment for tier |
 | `POST` | `/step` | `{"action": "FLAG(task_01, \"evidence\")"}` → Execute action |
-| `GET` | `/state` | Current observation + live hallucination stats |
+| `GET` | `/state` | Runtime-aware observation payload + build/container metadata |
+| `GET` | `/logs` | Structured logs summary (status/path counters + recent requests) |
 | `GET` | `/history` | Full action/reward timeline (JSON array) |
 | `GET` | `/dashboard` | Live visualization UI |
 
 ---
 
+## HF Spaces Health and Logs
+
+The Space now exposes two `200 OK` observability endpoints intended for build/runtime diagnostics:
+
+- `GET /state` returns:
+  - `status`
+  - `build` metadata (`container_id`, `build_id`, `git_sha`, `started_at`)
+  - current environment `observation`
+- `GET /logs` returns:
+  - `status`
+  - `build` metadata
+  - aggregate `totals`, `statuses`, and `paths`
+  - recent request `entries` with `method`, `path`, `status_code`, and `duration_ms`
+
+Quick check:
+
+```bash
+python scripts.py
+```
+
+---
+
+## OpenEnv Required Files
+
+OpenEnv validation expects a root-level `client.py`. This repository now provides:
+
+- `client.py` as the canonical OpenEnv evaluator script
+- `inference.py` as a compatibility wrapper for legacy commands
+
+Preferred command:
+
+```bash
+python client.py
+```
+
+---
+
 ## 🧪 Testing
 
 ```bash

client.py

@@ -0,0 +1,154 @@
+"""Client library for MissionCtrl environment.
+
+Provides HTTP client functions for interacting with the MissionCtrl environment API.
+Contains environment payload structures and example scenarios.
+"""
+
+import os
+from typing import Any, Dict, Optional
+
+import httpx
+from dotenv import load_dotenv
+
+# ---------------------------------------------------------------------------
+# Load .env file automatically
+# ---------------------------------------------------------------------------
+load_dotenv()
+
+# ---------------------------------------------------------------------------
+# Environment configuration
+# ---------------------------------------------------------------------------
+ENV_BASE_URL: str = os.environ.get("ENV_BASE_URL", "http://localhost:7860")
+
+# Known agent types in the environment
+KNOWN_AGENTS = (
+    "PlannerAgent",
+    "ResearchAgent",
+    "CoderAgent",
+    "TesterAgent",
+    "CommAgent",
+)
+
+# Available task tiers
+TASKS = ["easy", "medium", "hard", "special"]
+
+# Default max steps per episode
+MAX_STEPS = 5
+
+
+# ---------------------------------------------------------------------------
+# HTTP Client
+# ---------------------------------------------------------------------------
+http = httpx.Client(timeout=60.0)
+
+
+# ---------------------------------------------------------------------------
+# Environment API Functions
+# ---------------------------------------------------------------------------
+def reset_env(task_id: str, seed: Optional[int] = None) -> Dict[str, Any]:
+    """Reset the environment for a specific task.
+
+    Args:
+        task_id: The task tier to run (easy, medium, hard, special)
+        seed: Optional random seed for reproducibility
+
+    Returns:
+        Dictionary containing the initial observation
+    """
+    payload = {"task_id": task_id}
+    if seed is not None:
+        payload["seed"] = seed
+
+    resp = http.post(f"{ENV_BASE_URL}/reset", json=payload)
+    resp.raise_for_status()
+    return resp.json()
+
+
+def step_env(action: str) -> Dict[str, Any]:
+    """Execute one action in the environment.
+
+    Args:
+        action: The action string to execute (e.g., "APPROVE(task_1)")
+
+    Returns:
+        Dictionary containing the new observation, reward, done flag, and info
+    """
+    resp = http.post(f"{ENV_BASE_URL}/step", json={"action": action})
+    resp.raise_for_status()
+    return resp.json()
+
+
+def get_state() -> Dict[str, Any]:
+    """Get the current environment state (read-only).
+
+    Returns:
+        Dictionary containing the current observation
+    """
+    resp = http.get(f"{ENV_BASE_URL}/state")
+    resp.raise_for_status()
+    return resp.json()
+
+
+def get_history() -> list:
+    """Get the action history for the current episode.
+
+    Returns:
+        List of past actions and their results
+    """
+    resp = http.get(f"{ENV_BASE_URL}/history")
+    resp.raise_for_status()
+    return resp.json()
+
+
+def record_result(tier: str, score: float, steps: int, history: list, 
+                 score_breakdown: Optional[Dict] = None, 
+                 hallucination_stats: Optional[Dict] = None) -> Dict[str, str]:
+    """Push a completed episode result to the dashboard.
+
+    Args:
+        tier: Task tier (easy, medium, hard, special)
+        score: Final score for the episode
+        steps: Number of steps taken
+        history: Action history for the episode
+        score_breakdown: Optional detailed score breakdown
+        hallucination_stats: Optional hallucination detection statistics
+
+    Returns:
+        Confirmation response
+    """
+    payload = {
+        "tier": tier,
+        "score": score,
+        "steps": steps,
+        "history": history,
+        "score_breakdown": score_breakdown or {},
+        "hallucination_stats": hallucination_stats or {},
+    }
+    resp = http.post(f"{ENV_BASE_URL}/record", json=payload)
+    resp.raise_for_status()
+    return resp.json()
+
+
+# ---------------------------------------------------------------------------
+# Example Usage
+# ---------------------------------------------------------------------------
+def example_basic_usage():
+    """Example showing basic environment interaction."""
+    # Reset environment for easy task
+    result = reset_env("easy")
+    obs = result["observation"]
+    print(f"Started task with {len(obs['tasks'])} tasks")
+
+    # Take a step
+    step_result = step_env("NOOP")
+    print(f"Reward: {step_result['reward']}, Done: {step_result['done']}")
+
+    # Get current state
+    state = get_state()
+    print(f"Current step: {state['time_step']}")
+
+
+if __name__ == "__main__":
+    # For backward compatibility, delegate to inference script
+    from inference import main
+    main()

inference.py

@@ -46,7 +46,7 @@
 MODEL_NAME: str   = os.environ.get("MODEL_NAME", "openai/gpt-oss-120b")
 HF_TOKEN: str     = os.environ.get("HF_TOKEN", "")
 
-ENV_BASE_URL: str = os.environ.get("ENV_BASE_URL", "http://localhost:8000")
+ENV_BASE_URL: str = os.environ.get("ENV_BASE_URL", "http://localhost:7860")
 
 MAX_STEPS: int = int(os.environ.get("MAX_STEPS", "5"))
 TASKS: List[str] = ["easy", "medium", "hard", "special"]
@@ -100,7 +100,6 @@ class PromptTooLargeError(RuntimeError):
     """Raised when provider rejects a request as permanently oversized."""
 
 
-
 def _append_bounded_unique(bucket: List[str], value: str, limit: int) -> None:
     value = value.strip()
     if not value: