Add GAIA Unit 4 workbench

.gitignore

@@ -0,0 +1,6 @@
+.venv/
+__pycache__/
+*.pyc
+.env
+answers.local.json
+data/

README.md

@@ -1,15 +1,61 @@
 ---
-title: Template Final Assignment
-emoji: 🕵🏻‍♂️
-colorFrom: indigo
-colorTo: indigo
+title: GAIA Unit 4 Workbench
 sdk: gradio
-sdk_version: 5.25.2
 app_file: app.py
 pinned: false
-hf_oauth: true
-# optional, default duration is 8 hours/480 minutes. Max duration is 30 days/43200 minutes.
-hf_oauth_expiration_minutes: 480
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Hugging Face GAIA Agent Workbench
+
+Small Hugging Face Space/workbench for the Agents Course Unit 4 final assignment.
+
+## What It Does
+
+- Fetches the 20 filtered GAIA Level 1 questions from the course API.
+- Downloads task files into `data/files/`.
+- Lets you maintain exact-match answers in `answers.local.json`.
+- Submits answers to the course scoring API from the Gradio UI or CLI.
+
+## Required For Final Submission
+
+- Hugging Face account.
+- Public Hugging Face Space with this code.
+- HF login in the Space UI.
+- Exact final answers without `FINAL ANSWER` prefix.
+
+## Local Setup
+
+```bash
+python -m venv .venv
+source .venv/bin/activate
+pip install -r requirements.txt
+python gaia_workbench.py fetch
+python app.py
+```
+
+## CLI
+
+```bash
+python gaia_workbench.py fetch
+python gaia_workbench.py download-files
+python gaia_workbench.py status
+python gaia_workbench.py submit --username YOUR_HF_USERNAME --agent-code https://huggingface.co/spaces/YOUR_USERNAME/YOUR_SPACE/tree/main
+```
+
+## Answer File
+
+Create `answers.local.json` from the template:
+
+```bash
+cp answers.template.json answers.local.json
+```
+
+Then fill:
+
+```json
+{
+  "task-id": "exact answer"
+}
+```
+
+`answers.local.json` is gitignored because it may contain your final submission answers.

answers.template.json

@@ -0,0 +1 @@
+{}

app.py

@@ -1,196 +1,127 @@
+from __future__ import annotations
+
 import os
+
 import gradio as gr
-import requests
-import inspect
 import pandas as pd
 
-# (Keep Constants as is)
-# --- Constants ---
-DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
-
-# --- Basic Agent Definition ---
-# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
-class BasicAgent:
-    def __init__(self):
-        print("BasicAgent initialized.")
-    def __call__(self, question: str) -> str:
-        print(f"Agent received question (first 50 chars): {question[:50]}...")
-        fixed_answer = "This is a default answer."
-        print(f"Agent returning fixed answer: {fixed_answer}")
-        return fixed_answer
-
-def run_and_submit_all( profile: gr.OAuthProfile | None):
-    """
-    Fetches all questions, runs the BasicAgent on them, submits all answers,
-    and displays the results.
-    """
-    # --- Determine HF Space Runtime URL and Repo URL ---
-    space_id = os.getenv("SPACE_ID") # Get the SPACE_ID for sending link to the code
-
-    if profile:
-        username= f"{profile.username}"
-        print(f"User logged in: {username}")
-    else:
-        print("User not logged in.")
-        return "Please Login to Hugging Face with the button.", None
-
-    api_url = DEFAULT_API_URL
-    questions_url = f"{api_url}/questions"
-    submit_url = f"{api_url}/submit"
-
-    # 1. Instantiate Agent ( modify this part to create your agent)
-    try:
-        agent = BasicAgent()
-    except Exception as e:
-        print(f"Error instantiating agent: {e}")
-        return f"Error initializing agent: {e}", None
-    # In the case of an app running as a hugging Face space, this link points toward your codebase ( usefull for others so please keep it public)
-    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
-    print(agent_code)
-
-    # 2. Fetch Questions
-    print(f"Fetching questions from: {questions_url}")
-    try:
-        response = requests.get(questions_url, timeout=15)
-        response.raise_for_status()
-        questions_data = response.json()
-        if not questions_data:
-             print("Fetched questions list is empty.")
-             return "Fetched questions list is empty or invalid format.", None
-        print(f"Fetched {len(questions_data)} questions.")
-    except requests.exceptions.RequestException as e:
-        print(f"Error fetching questions: {e}")
-        return f"Error fetching questions: {e}", None
-    except requests.exceptions.JSONDecodeError as e:
-         print(f"Error decoding JSON response from questions endpoint: {e}")
-         print(f"Response text: {response.text[:500]}")
-         return f"Error decoding server response for questions: {e}", None
-    except Exception as e:
-        print(f"An unexpected error occurred fetching questions: {e}")
-        return f"An unexpected error occurred fetching questions: {e}", None
-
-    # 3. Run your Agent
-    results_log = []
-    answers_payload = []
-    print(f"Running agent on {len(questions_data)} questions...")
-    for item in questions_data:
-        task_id = item.get("task_id")
-        question_text = item.get("question")
-        if not task_id or question_text is None:
-            print(f"Skipping item with missing task_id or question: {item}")
-            continue
-        try:
-            submitted_answer = agent(question_text)
-            answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
-            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
-        except Exception as e:
-             print(f"Error running agent on task {task_id}: {e}")
-             results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
-
-    if not answers_payload:
-        print("Agent did not produce any answers to submit.")
-        return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
-
-    # 4. Prepare Submission 
-    submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload}
-    status_update = f"Agent finished. Submitting {len(answers_payload)} answers for user '{username}'..."
-    print(status_update)
-
-    # 5. Submit
-    print(f"Submitting {len(answers_payload)} answers to: {submit_url}")
+from gaia_workbench import (
+    API_URL,
+    answer_payload,
+    download_files,
+    fetch_questions,
+    load_answers,
+    load_questions,
+    status_rows,
+    submit_answers,
+    write_answer,
+)
+
+
+def refresh_questions() -> tuple[str, pd.DataFrame]:
+    questions = fetch_questions()
+    return f"Fetched {len(questions)} questions from {API_URL}", pd.DataFrame(status_rows())
+
+
+def refresh_status() -> pd.DataFrame:
+    return pd.DataFrame(status_rows())
+
+
+def save_answer(task_id: str, answer: str) -> tuple[str, pd.DataFrame]:
+    if not task_id.strip():
+        return "Missing task_id", refresh_status()
+    if not answer.strip():
+        return "Missing answer", refresh_status()
+    write_answer(task_id.strip(), answer.strip())
+    return f"Saved answer for {task_id.strip()}", refresh_status()
+
+
+def download_attached_files() -> str:
+    downloaded, failed = download_files()
+    parts = []
+    if downloaded:
+        parts.append("Downloaded:\n" + "\n".join(str(path) for path in downloaded))
+    if failed:
+        parts.append("Failed:\n" + "\n".join(failed))
+    return "\n\n".join(parts) if parts else "No attached files found."
+
+
+def _default_agent_code() -> str:
+    space_id = os.getenv("SPACE_ID", "").strip()
+    return f"https://huggingface.co/spaces/{space_id}/tree/main" if space_id else ""
+
+
+def _submit_with(username: str, agent_code: str) -> tuple[str, pd.DataFrame]:
+    agent_code = agent_code.strip() or _default_agent_code()
+
     try:
-        response = requests.post(submit_url, json=submission_data, timeout=60)
-        response.raise_for_status()
-        result_data = response.json()
-        final_status = (
-            f"Submission Successful!\n"
-            f"User: {result_data.get('username')}\n"
-            f"Overall Score: {result_data.get('score', 'N/A')}% "
-            f"({result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')} correct)\n"
-            f"Message: {result_data.get('message', 'No message received.')}"
-        )
-        print("Submission successful.")
-        results_df = pd.DataFrame(results_log)
-        return final_status, results_df
-    except requests.exceptions.HTTPError as e:
-        error_detail = f"Server responded with status {e.response.status_code}."
-        try:
-            error_json = e.response.json()
-            error_detail += f" Detail: {error_json.get('detail', e.response.text)}"
-        except requests.exceptions.JSONDecodeError:
-            error_detail += f" Response: {e.response.text[:500]}"
-        status_message = f"Submission Failed: {error_detail}"
-        print(status_message)
-        results_df = pd.DataFrame(results_log)
-        return status_message, results_df
-    except requests.exceptions.Timeout:
-        status_message = "Submission Failed: The request timed out."
-        print(status_message)
-        results_df = pd.DataFrame(results_log)
-        return status_message, results_df
-    except requests.exceptions.RequestException as e:
-        status_message = f"Submission Failed: Network error - {e}"
-        print(status_message)
-        results_df = pd.DataFrame(results_log)
-        return status_message, results_df
-    except Exception as e:
-        status_message = f"An unexpected error occurred during submission: {e}"
-        print(status_message)
-        results_df = pd.DataFrame(results_log)
-        return status_message, results_df
-
-
-# --- Build Gradio Interface using Blocks ---
-with gr.Blocks() as demo:
-    gr.Markdown("# Basic Agent Evaluation Runner")
+        result = submit_answers(username=username, agent_code=agent_code)
+    except Exception as exc:
+        return f"Submission failed: {exc}", refresh_status()
+
+    return (
+        "Submission successful\n"
+        f"User: {result.get('username')}\n"
+        f"Score: {result.get('score')}%\n"
+        f"Correct: {result.get('correct_count')}/{result.get('total_attempted')}\n"
+        f"Message: {result.get('message')}",
+        refresh_status(),
+    )
+
+
+def submit_logged_in(profile: gr.OAuthProfile | None) -> tuple[str, pd.DataFrame]:
+    if not profile:
+        return "Please login with Hugging Face first, or use manual submit.", refresh_status()
+    return _submit_with(profile.username, _default_agent_code())
+
+
+def submit_manual(manual_username: str, manual_agent_code: str) -> tuple[str, pd.DataFrame]:
+    return _submit_with(manual_username.strip(), manual_agent_code.strip())
+
+
+with gr.Blocks(title="GAIA Unit 4 Workbench") as demo:
+    gr.Markdown("# GAIA Unit 4 Workbench")
     gr.Markdown(
-        """
-        **Instructions:**
-
-        1.  Please clone this space, then modify the code to define your agent's logic, the tools, the necessary packages, etc ...
-        2.  Log in to your Hugging Face account using the button below. This uses your HF username for submission.
-        3.  Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.
-
-        ---
-        **Disclaimers:**
-        Once clicking on the "submit button, it can take quite some time ( this is the time for the agent to go through all the questions).
-        This space provides a basic setup and is intentionally sub-optimal to encourage you to develop your own, more robust solution. For instance for the delay process of the submit button, a solution could be to cache the answers and submit in a seperate action or even to answer the questions in async.
-        """
+        "Fetch the course questions, fill exact-match answers, then submit. "
+        "On Hugging Face Spaces, login with the button before submission."
     )
 
-    gr.LoginButton()
+    is_space_runtime = bool(os.getenv("SPACE_ID", "").strip())
+    if is_space_runtime:
+        gr.LoginButton()
 
-    run_button = gr.Button("Run Evaluation & Submit All Answers")
+    with gr.Row():
+        fetch_button = gr.Button("Fetch Questions")
+        download_button = gr.Button("Download Attached Files")
+        refresh_button = gr.Button("Refresh Status")
 
-    status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False)
-    # Removed max_rows=10 from DataFrame constructor
-    results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
+    status = gr.Textbox(label="Status", lines=5)
+    table = gr.DataFrame(label="Questions / Answers", wrap=True, value=pd.DataFrame(status_rows()))
 
-    run_button.click(
-        fn=run_and_submit_all,
-        outputs=[status_output, results_table]
+    with gr.Row():
+        task_id_input = gr.Textbox(label="Task ID")
+        answer_input = gr.Textbox(label="Exact Answer")
+    save_button = gr.Button("Save Answer")
+
+    gr.Markdown("## Submit")
+    manual_username = gr.Textbox(label="HF Username fallback", placeholder="Used only when not logged in via HF OAuth")
+    manual_agent_code = gr.Textbox(
+        label="Agent Code URL fallback",
+        placeholder="https://huggingface.co/spaces/YOUR_USERNAME/YOUR_SPACE/tree/main",
     )
+    with gr.Row():
+        submit_login_button = gr.Button("Submit With HF Login", visible=is_space_runtime)
+        submit_manual_button = gr.Button("Submit Manual")
+
+    fetch_button.click(fn=refresh_questions, outputs=[status, table])
+    download_button.click(fn=download_attached_files, outputs=status)
+    refresh_button.click(fn=refresh_status, outputs=table)
+    save_button.click(fn=save_answer, inputs=[task_id_input, answer_input], outputs=[status, table])
+    submit_login_button.click(fn=submit_logged_in, outputs=[status, table])
+    submit_manual_button.click(fn=submit_manual, inputs=[manual_username, manual_agent_code], outputs=[status, table])
+
 
 if __name__ == "__main__":
-    print("\n" + "-"*30 + " App Starting " + "-"*30)
-    # Check for SPACE_HOST and SPACE_ID at startup for information
-    space_host_startup = os.getenv("SPACE_HOST")
-    space_id_startup = os.getenv("SPACE_ID") # Get SPACE_ID at startup
-
-    if space_host_startup:
-        print(f"✅ SPACE_HOST found: {space_host_startup}")
-        print(f"   Runtime URL should be: https://{space_host_startup}.hf.space")
-    else:
-        print("ℹ️  SPACE_HOST environment variable not found (running locally?).")
-
-    if space_id_startup: # Print repo URLs if SPACE_ID is found
-        print(f"✅ SPACE_ID found: {space_id_startup}")
-        print(f"   Repo URL: https://huggingface.co/spaces/{space_id_startup}")
-        print(f"   Repo Tree URL: https://huggingface.co/spaces/{space_id_startup}/tree/main")
-    else:
-        print("ℹ️  SPACE_ID environment variable not found (running locally?). Repo URL cannot be determined.")
-
-    print("-"*(60 + len(" App Starting ")) + "\n")
-
-    print("Launching Gradio Interface for Basic Agent Evaluation...")
-    demo.launch(debug=True, share=False)
+    print(f"Loaded {len(load_questions())} questions and {len(load_answers())} local answers.")
+    print(f"Prepared {len(answer_payload())} answers for submission.")
+    demo.launch()

deploy_to_hf.py

@@ -0,0 +1,37 @@
+from __future__ import annotations
+
+import getpass
+from pathlib import Path
+
+from huggingface_hub import HfApi
+
+
+ROOT = Path(__file__).resolve().parent
+REPO_ID = "smashburger-dev/Final_Assignment_Template"
+
+
+def main() -> None:
+    token = getpass.getpass("HF token: ").strip()
+    if not token:
+        raise SystemExit("Missing token")
+
+    api = HfApi(token=token)
+    api.upload_folder(
+        repo_id=REPO_ID,
+        repo_type="space",
+        folder_path=str(ROOT),
+        ignore_patterns=[
+            ".venv/**",
+            "__pycache__/**",
+            "data/**",
+            "answers.local.json",
+            "*.pyc",
+            ".env",
+        ],
+        commit_message="Add GAIA Unit 4 workbench",
+    )
+    print(f"Uploaded to https://huggingface.co/spaces/{REPO_ID}/tree/main")
+
+
+if __name__ == "__main__":
+    main()

gaia_workbench.py

@@ -0,0 +1,186 @@
+from __future__ import annotations
+
+import argparse
+import json
+import os
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Any
+
+import requests
+
+
+API_URL = "https://agents-course-unit4-scoring.hf.space"
+ROOT = Path(__file__).resolve().parent
+DATA_DIR = ROOT / "data"
+QUESTIONS_PATH = DATA_DIR / "questions.json"
+FILES_DIR = DATA_DIR / "files"
+ANSWERS_PATH = ROOT / "answers.local.json"
+
+
+@dataclass(frozen=True)
+class Question:
+    task_id: str
+    question: str
+    file_name: str = ""
+
+
+def _request_json(method: str, url: str, **kwargs: Any) -> Any:
+    response = requests.request(method, url, timeout=60, **kwargs)
+    response.raise_for_status()
+    return response.json()
+
+
+def fetch_questions() -> list[Question]:
+    DATA_DIR.mkdir(parents=True, exist_ok=True)
+    raw_questions = _request_json("GET", f"{API_URL}/questions")
+    QUESTIONS_PATH.write_text(json.dumps(raw_questions, indent=2, ensure_ascii=False), encoding="utf-8")
+    return [
+        Question(
+            task_id=str(item["task_id"]),
+            question=str(item["question"]),
+            file_name=str(item.get("file_name") or ""),
+        )
+        for item in raw_questions
+    ]
+
+
+def load_questions() -> list[Question]:
+    if not QUESTIONS_PATH.exists():
+        return fetch_questions()
+    raw_questions = json.loads(QUESTIONS_PATH.read_text(encoding="utf-8"))
+    return [
+        Question(
+            task_id=str(item["task_id"]),
+            question=str(item["question"]),
+            file_name=str(item.get("file_name") or ""),
+        )
+        for item in raw_questions
+    ]
+
+
+def load_answers() -> dict[str, str]:
+    if not ANSWERS_PATH.exists():
+        return {}
+    raw_answers = json.loads(ANSWERS_PATH.read_text(encoding="utf-8"))
+    return {str(task_id): str(answer).strip() for task_id, answer in raw_answers.items() if str(answer).strip()}
+
+
+def write_answer(task_id: str, answer: str) -> None:
+    answers = load_answers()
+    answers[task_id] = answer.strip()
+    ANSWERS_PATH.write_text(json.dumps(answers, indent=2, ensure_ascii=False) + "\n", encoding="utf-8")
+
+
+def download_task_file(task_id: str, file_name: str = "") -> Path:
+    FILES_DIR.mkdir(parents=True, exist_ok=True)
+    response = requests.get(f"{API_URL}/files/{task_id}", timeout=60)
+    response.raise_for_status()
+    target_name = file_name or task_id
+    target = FILES_DIR / target_name
+    target.write_bytes(response.content)
+    return target
+
+
+def download_files() -> tuple[list[Path], list[str]]:
+    downloaded: list[Path] = []
+    failed: list[str] = []
+    for question in load_questions():
+        if question.file_name:
+            try:
+                downloaded.append(download_task_file(question.task_id, question.file_name))
+            except requests.HTTPError as exc:
+                failed.append(f"{question.task_id} ({question.file_name}): {exc}")
+    return downloaded, failed
+
+
+def answer_payload() -> list[dict[str, str]]:
+    answers = load_answers()
+    questions = load_questions()
+    return [
+        {"task_id": question.task_id, "submitted_answer": answers[question.task_id]}
+        for question in questions
+        if question.task_id in answers
+    ]
+
+
+def submit_answers(username: str, agent_code: str) -> dict[str, Any]:
+    payload = {
+        "username": username.strip(),
+        "agent_code": agent_code.strip(),
+        "answers": answer_payload(),
+    }
+    if not payload["username"]:
+        raise ValueError("Missing username")
+    if not payload["agent_code"]:
+        raise ValueError("Missing agent_code")
+    if not payload["answers"]:
+        raise ValueError("No answers found in answers.local.json")
+    return _request_json("POST", f"{API_URL}/submit", json=payload)
+
+
+def status_rows() -> list[dict[str, str]]:
+    answers = load_answers()
+    rows = []
+    for index, question in enumerate(load_questions(), start=1):
+        rows.append(
+            {
+                "#": str(index),
+                "task_id": question.task_id,
+                "file": question.file_name,
+                "answered": "yes" if question.task_id in answers else "no",
+                "answer": answers.get(question.task_id, ""),
+                "question": question.question,
+            }
+        )
+    return rows
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="Hugging Face GAIA Unit 4 workbench")
+    subparsers = parser.add_subparsers(dest="command", required=True)
+
+    subparsers.add_parser("fetch", help="Fetch and cache questions")
+    subparsers.add_parser("download-files", help="Download all attached task files")
+    subparsers.add_parser("status", help="Print answer coverage")
+
+    set_answer_parser = subparsers.add_parser("set-answer", help="Set one exact-match answer")
+    set_answer_parser.add_argument("task_id")
+    set_answer_parser.add_argument("answer")
+
+    submit_parser = subparsers.add_parser("submit", help="Submit filled local answers")
+    submit_parser.add_argument("--username", default=os.getenv("HF_USERNAME", ""))
+    submit_parser.add_argument("--agent-code", default=os.getenv("AGENT_CODE", ""))
+
+    args = parser.parse_args()
+
+    if args.command == "fetch":
+        questions = fetch_questions()
+        print(f"Fetched {len(questions)} questions -> {QUESTIONS_PATH}")
+    elif args.command == "download-files":
+        downloaded, failed = download_files()
+        print(f"Downloaded {len(downloaded)} files")
+        for path in downloaded:
+            print(path)
+        if failed:
+            print(f"Failed {len(failed)} files")
+            for item in failed:
+                print(item)
+    elif args.command == "status":
+        rows = status_rows()
+        answered_count = sum(1 for row in rows if row["answered"] == "yes")
+        print(f"Answered {answered_count}/{len(rows)}")
+        for row in rows:
+            marker = "[x]" if row["answered"] == "yes" else "[ ]"
+            file_part = f" file={row['file']}" if row["file"] else ""
+            print(f"{marker} {row['#']}. {row['task_id']}{file_part} - {row['question']}")
+    elif args.command == "set-answer":
+        write_answer(args.task_id, args.answer)
+        print(f"Saved answer for {args.task_id}")
+    elif args.command == "submit":
+        result = submit_answers(args.username, args.agent_code)
+        print(json.dumps(result, indent=2, ensure_ascii=False))
+
+
+if __name__ == "__main__":
+    main()

requirements.txt

@@ -1,2 +1,5 @@
-gradio
-requests
+gradio[oauth]>=4.44.0
+huggingface_hub<1.0
+pandas>=2.2.0
+openpyxl>=3.1.0
+requests>=2.32.0