Update agent code

agent.py

@@ -0,0 +1,129 @@
+"""GAIA benchmark agent using *smolagents*.
+
+This module exposes:
+
+* ``gaia_agent()`` – factory returning a ready‑to‑use agent instance.
+* ``GAIAAgent``  – subclass of ``smolagents.CodeAgent``.
+
+The LLM backend is chosen at runtime via the ``MODEL_PROVIDER``
+environment variable (``hf`` or ``openai``) exactly like *example.py*.
+"""
+
+import os
+from typing import Any, Sequence
+
+from dotenv import load_dotenv
+
+# SmolAgents Tools
+from smolagents import (
+    CodeAgent,
+    DuckDuckGoSearchTool,
+    Tool
+)
+
+# Custom Tools from tools.py
+from tools import (
+    PythonRunTool,
+    ExcelLoaderTool,
+    YouTubeTranscriptTool,
+    AudioTranscriptionTool,
+    SimpleOCRTool,
+)
+
+ 
+# ---------------------------------------------------------------------------
+# Load the added system prompt from system_prompt.txt (located in the same directory)
+# ---------------------------------------------------------------------------
+ADDED_PROMPT_PATH = os.path.join(os.path.dirname(__file__), "added_prompt.txt")
+with open(ADDED_PROMPT_PATH, "r", encoding="utf-8") as f:
+    ADDED_PROMPT = f.read().strip()
+
+
+# ---------------------------------------------------------------------------
+# Model selection helper
+# ---------------------------------------------------------------------------
+
+load_dotenv()  # Make sure we read credentials from .env when running locally
+
+def _select_model():
+    """Return a smolagents *model* as configured by the ``MODEL_PROVIDER`` env."""
+
+    provider = os.getenv("MODEL_PROVIDER", "hf").lower()
+
+    if provider == "hf":
+        from smolagents import InferenceClientModel
+        hf_model_id = os.getenv("HF_MODEL", "HuggingFaceH4/zephyr-7b-beta")
+        hf_token = os.getenv("HF_API_KEY")
+        return InferenceClientModel(
+            model_id=hf_model_id,
+            token=hf_token
+        )
+
+    if provider == "openai":
+        from smolagents import OpenAIServerModel
+        openai_model_id = os.getenv("OPENAI_MODEL", "gpt-3.5-turbo")
+        openai_token = os.getenv("OPENAI_API_KEY")
+        return OpenAIServerModel(
+            model_id=openai_model_id,
+            api_key=openai_token
+        )
+
+    raise ValueError(
+        f"Unsupported MODEL_PROVIDER: {provider!r}. "
+        "Use 'hf' (default) or 'openai'."
+    )
+
+# ---------------------------------------------------------------------------
+# Core Agent implementation
+# ---------------------------------------------------------------------------
+
+DEFAULT_TOOLS = [
+    DuckDuckGoSearchTool(),
+    PythonRunTool(),
+    ExcelLoaderTool(),
+    YouTubeTranscriptTool(),
+    AudioTranscriptionTool(),
+    SimpleOCRTool(),
+]
+
+class GAIAAgent(CodeAgent):
+    def __init__(
+        self,
+        tools=None
+    ):
+        super().__init__(
+            tools=tools or DEFAULT_TOOLS,
+            model=_select_model()
+        )
+        # Append the additional prompt to the existing system prompt
+        self.prompt_templates["system_prompt"] += f"\n\n{ADDED_PROMPT}"
+
+    # Convenience so the object itself can be *called* directly
+    def __call__(self, question: str, **kwargs: Any) -> str:
+        steps = self.run(question, **kwargs)
+        # If steps is a primitive, just return it
+        if isinstance(steps, (int, float, str)):
+            return str(steps).strip()
+        last_step = None
+        for step in steps:
+            last_step = step
+        # Defensive: handle int/float/str directly
+        if isinstance(last_step, (int, float, str)):
+            return str(last_step).strip()
+        answer = getattr(last_step, "answer", None)
+        if answer is not None:
+            return str(answer).strip()
+        return str(last_step).strip()
+
+# ---------------------------------------------------------------------------
+# Factory helpers expected by app.py
+# ---------------------------------------------------------------------------
+
+def gaia_agent(*, extra_tools: Sequence[Tool] | None = None) -> GAIAAgent:
+    # Compose the toolset: always include all default tools, plus any extras
+    toolset = list(DEFAULT_TOOLS)
+    if extra_tools:
+        toolset.extend(extra_tools)
+    return GAIAAgent(tools=toolset)
+
+__all__ = ["GAIAAgent", "gaia_agent"]

app.py

@@ -1,196 +1,199 @@
-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}")
-    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")
-    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.
-        """
-    )
-
-    gr.LoginButton()
-
-    run_button = gr.Button("Run Evaluation & Submit All Answers")
-
-    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)
-
-    run_button.click(
-        fn=run_and_submit_all,
-        outputs=[status_output, results_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)
+import os
+import gradio as gr
+import requests
+import pandas as pd
+
+# --- Our Agent ---
+from agent import gaia_agent
+
+# Debugging level. If DEBUG=0 then DEBUG will be False. If DEBUG=1 then DEBUG will be True.
+DEBUG = os.getenv("DEBUG", "0") == "1"
+
+# (Keep Constants as is)
+# --- Constants ---
+DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
+
+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 (now using smolagents)
+    try:
+        agent = gaia_agent()
+        print("SmolAgent instantiated successfully.")
+    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 ( useful 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}")
+    import json
+
+    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 json.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 requests.exceptions.RequestException as e:
+        print(f"Error fetching questions: {e}")
+        return f"Error fetching 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 the 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)
+            # --- DEBUG LOGGING ---
+            if DEBUG:
+                print(f"[DEBUG] Task {task_id}: Answer type: {type(submitted_answer)}, Value: {repr(submitted_answer)}")
+            else:
+                print(f"[{task_id}] {question_text[:50]}... → {submitted_answer[:40]}")
+
+            # Force string type here just in case (defensive)
+            submitted_answer = str(submitted_answer).strip()
+            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}")
+    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("# Agent Evaluation Runner")
+    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.
+        """
+    )
+
+    gr.LoginButton()
+
+    run_button = gr.Button("Run Evaluation & Submit All Answers")
+
+    status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False)
+    results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
+
+    run_button.click(
+        fn=run_and_submit_all,
+        outputs=[status_output, results_table]
+    )
+
+if __name__ == "__main__":
+    print("\n" + "-"*30 + " App Starting " + "-"*30)
+    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 Agent Evaluation…")
+    demo.launch(debug=True, share=False)

requirements.txt

@@ -1,2 +1,8 @@
-gradio
-requests
+gradio
+requests
+pandas
+smolagents[openai]
+duckduckgo-search
+youtube-transcript-api
+pytesseract
+pillow

tools.py

@@ -0,0 +1,170 @@
+# Custom tools for smolagents GAIA agent
+from __future__ import annotations
+import contextlib
+import io
+import os
+from typing import Any, Dict, List
+
+from smolagents import Tool
+
+# ---- 1. PythonRunTool ------------------------------------------------------
+class PythonRunTool(Tool):
+    name = "python_run"
+    description = """
+        Execute trusted Python code and return printed output + repr() of the last expression (or _result variable).
+    """
+    inputs = {
+        "code": {
+            "type": "string",
+            "description": "Python code to execute",
+            "required": True
+        }
+    }
+    output_type = "string"
+
+    def forward(self, code: str) -> str:
+        buf, ns = io.StringIO(), {}
+        last = None
+        try:
+            with contextlib.redirect_stdout(buf):
+                exec(compile(code, "<agent-python>", "exec"), {}, ns)
+            last = ns.get("_result", None)
+        except Exception as e:
+            raise RuntimeError(f"PythonRunTool error: {e}") from e
+        out = buf.getvalue()
+        # Always return a string
+        result = (out + (repr(last) if last is not None else "")).strip()
+        return str(result)
+
+# ---- 2. ExcelLoaderTool ----------------------------------------------------
+class ExcelLoaderTool(Tool):
+    name = "load_spreadsheet"
+    description = """
+        Read .xlsx/.xls/.csv from disk and return rows as a list of dictionaries with string keys.
+    """
+    inputs = {
+        "path": {
+            "type": "string",
+            "description": "Path to .csv/.xls/.xlsx file",
+            "required": True
+        },
+        "sheet": {
+            "type": "string",
+            "description": "Sheet name or index (optional, required for Excel files only)",
+            "required": False,
+            "default": "",
+            "nullable": True
+        }
+    }
+    output_type = "array"
+
+    def forward(self, path: str, sheet: str | int | None = None) -> str:
+        import pandas as pd
+        if not os.path.isfile(path):
+            raise FileNotFoundError(path)
+        ext = os.path.splitext(path)[1].lower()
+        if sheet == "":
+            sheet = None
+        if ext == ".csv":
+            df = pd.read_csv(path)
+        else:
+            df = pd.read_excel(path, sheet_name=sheet)
+        if isinstance(df, dict):
+            # If user did not specify a sheet, use the first one found
+            first_sheet = next(iter(df))
+            df = df[first_sheet]
+        records = [{str(k): v for k, v in row.items()} for row in df.to_dict(orient="records")]
+        # Always return a string
+        return str(records)
+
+# ---- 3. YouTubeTranscriptTool ---------------------------------------------
+class YouTubeTranscriptTool(Tool):
+    name = "youtube_transcript"
+    description = """
+        Return the subtitles of a YouTube URL using youtube-transcript-api.
+    """
+    inputs = {
+        "url": {
+            "type": "string",
+            "description": "YouTube URL",
+            "required": True
+        },
+        "lang": {
+            "type": "string",
+            "description": "Transcript language (default: en)",
+            "required": False,
+            "default": "en",
+            "nullable": True
+        }
+    }
+    output_type = "string"
+
+    def forward(self, url: str, lang: str = "en") -> str:
+        from urllib.parse import urlparse, parse_qs
+        from youtube_transcript_api._api import YouTubeTranscriptApi
+        vid = parse_qs(urlparse(url).query).get("v", [None])[0] or url.split("/")[-1]
+        data = YouTubeTranscriptApi.get_transcript(vid, languages=[lang, "en", "en-US", "en-GB"])
+        text = " ".join(d["text"] for d in data).strip()
+        return str(text)
+
+# ---- 4. AudioTranscriptionTool --------------------------------------------
+class AudioTranscriptionTool(Tool):
+    name = "transcribe_audio"
+    description = """
+        Transcribe an audio file with OpenAI Whisper, returns plain text."
+    """
+    inputs = {
+        "path": {
+            "type": "string",
+            "description": "Path to audio file",
+            "required": True
+        },
+        "model": {
+            "type": "string",
+            "description": "Model name for transcription (default: whisper-1)",
+            "required": False,
+            "default": "whisper-1",
+            "nullable": True
+        }
+    }
+    output_type = "string"
+
+    def forward(self, path: str, model: str = "whisper-1") -> str:
+        import openai
+        if not os.path.isfile(path):
+            raise FileNotFoundError(path)
+        client = openai.OpenAI(api_key=os.getenv("OPENAI_API_KEY"))
+        with open(path, "rb") as fp:
+            transcript = client.audio.transcriptions.create(model=model, file=fp)
+        return str(transcript.text.strip())
+
+# ---- 5. SimpleOCRTool ------------------------------------------------------
+class SimpleOCRTool(Tool):
+    name = "image_ocr"
+    description = """
+        Return any text spotted in an image via pytesseract OCR.
+    """
+    inputs = {
+        "path": {
+            "type": "string",
+            "description": "Path to image file",
+            "required": True
+        }
+    }
+    output_type = "string"
+
+    def forward(self, path: str) -> str:
+        from PIL import Image
+        import pytesseract
+        if not os.path.isfile(path):
+            raise FileNotFoundError(path)
+        return str(pytesseract.image_to_string(Image.open(path)).strip())
+
+# ---------------------------------------------------------------------------
+__all__ = [
+    "PythonRunTool",
+    "ExcelLoaderTool",
+    "YouTubeTranscriptTool",
+    "AudioTranscriptionTool",
+    "SimpleOCRTool",
+]