| import os |
| import gradio as gr |
| import requests |
| import inspect |
| import pandas as pd |
| import re |
| import warnings |
| warnings.filterwarnings("ignore") |
| import json |
| import logging |
| from typing import TypedDict, Annotated, Dict, Any |
| from json_repair import repair_json |
| import requests |
| from bs4 import BeautifulSoup |
| from pydantic import BaseModel, Field |
| from typing import Dict |
| from transformers import AutoTokenizer, AutoModelForCausalLM |
| import torch |
| from langgraph.graph import StateGraph, START, END |
| from langgraph.graph.message import add_messages |
| from langchain_core.messages import AnyMessage, HumanMessage, AIMessage |
| from langchain_community.retrievers import BM25Retriever |
| from langchain_core.tools import Tool |
| from langchain_core.documents import Document |
| from langgraph.prebuilt import ToolNode, tools_condition |
| from sentence_transformers import SentenceTransformer |
| from sklearn.metrics.pairwise import cosine_similarity |
| |
|
|
| |
| |
| DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" |
|
|
| sentence_transformer_model = SentenceTransformer("all-mpnet-base-v2") |
|
|
| logger = logging.getLogger("agent") |
| logging.basicConfig(level=logging.INFO) |
|
|
| class Config(object): |
| def __init__(self): |
| self.random_state = 42 |
| self.max_len = 256 |
| self.reasoning_max_len = 128 |
| self.temperature = 0.1 |
| self.DEVICE = "cuda" if torch.cuda.is_available() else "cpu" |
| self.model_name = "mistralai/Mistral-7B-Instruct-v0.2" |
| |
| |
| self.reasoning_model_name = "mistralai/Mistral-7B-Instruct-v0.2" |
|
|
|
|
| config = Config() |
|
|
| tokenizer = AutoTokenizer.from_pretrained(config.model_name) |
| model = AutoModelForCausalLM.from_pretrained( |
| config.model_name, |
| torch_dtype=torch.float16, |
| device_map="auto" |
| ) |
|
|
| |
| |
| |
| |
| |
| |
|
|
| def generate(prompt): |
| """ |
| Generate a text completion from a causal language model given a prompt. |
| |
| Parameters |
| ---------- |
| prompt : str |
| Input text prompt used to condition the language model. |
| |
| Returns |
| ------- |
| str |
| The generated continuation text, decoded into a string with special |
| tokens removed and leading/trailing whitespace stripped. |
| |
| """ |
| inputs = tokenizer(prompt, return_tensors="pt").to(model.device) |
|
|
| with torch.no_grad(): |
| outputs = model.generate( |
| **inputs, |
| max_new_tokens=config.max_len, |
| temperature=config.temperature, |
| ) |
|
|
| generated = outputs[0][inputs["input_ids"].shape[-1]:] |
|
|
| return tokenizer.decode(generated, skip_special_tokens=True).strip() |
|
|
| def reasoning_generate(prompt): |
| """ |
| Generate a text completion from a causal language model given a prompt. |
| |
| Parameters |
| ---------- |
| prompt : str |
| Input text prompt used to condition the language model. |
| |
| Returns |
| ------- |
| str |
| The generated continuation text, decoded into a string with special |
| tokens removed and leading/trailing whitespace stripped. |
| |
| """ |
| inputs = reasoning_tokenizer(prompt, return_tensors="pt").to(model.device) |
|
|
| with torch.no_grad(): |
| outputs = reasoning_model.generate( |
| **inputs, |
| max_new_tokens=config.reasoning_max_len, |
| temperature=config.temperature, |
| ) |
|
|
| generated = outputs[0][inputs["input_ids"].shape[-1]:] |
|
|
| return reasoning_tokenizer.decode(generated, skip_special_tokens=True).strip() |
|
|
| class Action(BaseModel): |
| tool: str = Field(...) |
| args: Dict |
|
|
| |
| class AgentState(TypedDict): |
| messages: Annotated[list[AnyMessage], add_messages] |
| proposed_action: str |
| information: str |
| output: str |
| confidence: float |
| judge_explanation: str |
|
|
| ALL_TOOLS = { |
| "web_search": ["query"], |
| "visit_webpage": ["url"], |
| } |
|
|
| ALLOWED_TOOLS = { |
| "web_search": ["query"], |
| "visit_webpage": ["url"], |
| } |
|
|
|
|
| def visit_webpage(url: str) -> str: |
| """ |
| Fetch and read the content of a webpage. |
| Args: |
| url: URL of the webpage |
| Returns: |
| Extracted readable text (truncated) |
| """ |
|
|
| headers = { |
| "User-Agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120 Safari/537.36" |
| } |
|
|
| response = requests.get(url, headers=headers, timeout=10) |
| response.raise_for_status() |
|
|
| soup = BeautifulSoup(response.text, "html.parser") |
|
|
| paragraphs = [p.get_text() for p in soup.find_all("p")] |
| text = "\n".join(paragraphs) |
|
|
| return (text[:500], text[500:1000]) |
|
|
|
|
| def visit_webpage(url: str) -> str: |
| headers = { |
| "User-Agent": "Mozilla/5.0" |
| } |
|
|
| response = requests.get(url, headers=headers, timeout=10) |
| response.raise_for_status() |
|
|
| soup = BeautifulSoup(response.text, "html.parser") |
|
|
| |
| for tag in soup(["script", "style"]): |
| tag.extract() |
|
|
| |
| elements = soup.find_all(["p", "dd"]) |
|
|
| text = " \n ".join(el.get_text(strip=False) for el in elements) |
|
|
| return (text[:1000], ) |
|
|
|
|
| def web_search(query: str, num_results: int = 10): |
| """ |
| Search the internet for the query provided |
| Args: |
| query: Query to search in the internet |
| Returns: |
| list of urls |
| """ |
|
|
| url = "https://html.duckduckgo.com/html/" |
| headers = { |
| "User-Agent": "Mozilla/5.0" |
| } |
|
|
| response = requests.post(url, data={"q": query}, headers=headers) |
| |
| soup = BeautifulSoup(response.text, "html.parser") |
| return [a.get("href") for a in soup.select(".result__a")[:num_results]] |
|
|
| def planner_node(state: AgentState): |
| """ |
| Planning node for a tool-using LLM agent. |
| |
| The planner enforces: |
| - Strict JSON-only output |
| - Tool selection constrained to predefined tools |
| - Argument generation limited to user-provided information |
| |
| Parameters |
| ---------- |
| state : dict |
| Agent state dictionary containing: |
| - "messages" (str): The user's natural language request. |
| |
| Returns |
| ------- |
| dict |
| Updated state dictionary with additional keys: |
| - "proposed_action" (dict): Parsed JSON tool call in the form: |
| { |
| "tool": "<tool_name>", |
| "args": {...} |
| } |
| - "risk_score" (float): Initialized risk score (default 0.0). |
| - "decision" (str): Initial decision ("allow" by default). |
| |
| Behavior |
| -------- |
| 1. Constructs a planning prompt including: |
| - Available tools and allowed arguments |
| - Strict JSON formatting requirements |
| - Example of valid output |
| 2. Calls the language model via `generate()`. |
| 3. Attempts to extract valid JSON from the model output. |
| 4. Repairs malformed JSON using `repair_json`. |
| 5. Stores the parsed action into the agent state. |
| |
| Security Notes |
| -------------- |
| - This node does not enforce tool-level authorization. |
| - It does not validate hallucinated tools. |
| - It does not perform risk scoring beyond initializing values. |
| - Downstream nodes must implement: |
| * Tool whitelist validation |
| * Argument validation |
| * Risk scoring and mitigation |
| * Execution authorization |
| |
| Intended Usage |
| -------------- |
| Designed for multi-agent or LangGraph-style workflows where: |
| Planner → Risk Assessment → Tool Executor → Logger |
| |
| This node represents the *planning layer* of the agent architecture. |
| """ |
|
|
| user_input = state["messages"][-1].content |
|
|
| prompt = f""" |
| You are a planning agent. |
| |
| You MUST return ONLY valid JSON as per the tools specs below ONLY. |
| No extra text. |
| DO NOT invent anything additional beyond the user request provided. Keep it strict to the user request information provided. The question and the query should be fully relevant to the user request provided, no deviation and hallucination. If possible and makes sense then the query should be exactly the user request. |
| |
| The available tools and their respective arguments are: {{ |
| "web_search": ["query"], |
| "visit_webpage": ["url"], |
| }} |
| |
| Return exactly the following format: |
| Response: |
| {{ |
| "tool": "...", |
| "args": {{...}} |
| }} |
| |
| User request: Who nominated the only Featured Article on English Wikipedia about a dinosaur that was promoted in November 2016?. Example of valid JSON expected: |
| Response: |
| {{"tool": "web_search", |
| "args": {{"query": "Who nominated the only Featured Article on English Wikipedia about a dinosaur that was promoted in November 2016?", |
| }} |
| }} |
| |
| Return only one Response! |
| |
| User request: |
| {user_input} |
| """ |
|
|
| output = generate(prompt) |
|
|
| state["proposed_action"] = output.split("Response:")[-1] |
| fixed = repair_json(state["proposed_action"]) |
| data = json.loads(fixed) |
| state["proposed_action"] = data |
|
|
| return state |
|
|
| def planner_node(state: AgentState): |
| """ |
| Planning node for a tool-using LLM agent. |
| |
| The planner enforces: |
| - Strict JSON-only output |
| - Tool selection constrained to predefined tools |
| - Argument generation limited to user-provided information |
| |
| Parameters |
| ---------- |
| state : dict |
| Agent state dictionary containing: |
| - "messages" (str): The user's natural language request. |
| |
| Returns |
| ------- |
| dict |
| Updated state dictionary with additional keys: |
| - "proposed_action" (dict): Parsed JSON tool call in the form: |
| { |
| "tool": "<tool_name>", |
| "args": {...} |
| } |
| - "risk_score" (float): Initialized risk score (default 0.0). |
| - "decision" (str): Initial decision ("allow" by default). |
| |
| Behavior |
| -------- |
| 1. Constructs a planning prompt including: |
| - Available tools and allowed arguments |
| - Strict JSON formatting requirements |
| - Example of valid output |
| 2. Calls the language model via `generate()`. |
| 3. Attempts to extract valid JSON from the model output. |
| 4. Repairs malformed JSON using `repair_json`. |
| 5. Stores the parsed action into the agent state. |
| |
| Security Notes |
| -------------- |
| - This node does not enforce tool-level authorization. |
| - It does not validate hallucinated tools. |
| - It does not perform risk scoring beyond initializing values. |
| - Downstream nodes must implement: |
| * Tool whitelist validation |
| * Argument validation |
| * Risk scoring and mitigation |
| * Execution authorization |
| |
| Intended Usage |
| -------------- |
| Designed for multi-agent or LangGraph-style workflows where: |
| Planner → Risk Assessment → Tool Executor → Logger |
| |
| This node represents the *planning layer* of the agent architecture. |
| """ |
|
|
| user_input = state["messages"][-1].content |
|
|
| prompt = f""" |
| You are a planning agent. |
| |
| You MUST return ONLY valid JSON as per the tools specs below ONLY. |
| No extra text. |
| DO NOT invent anything additional beyond the user request provided. Keep it strict to the user request information provided. The question and the query should be fully relevant to the user request provided, no deviation and hallucination. If possible and makes sense then the query should be exactly the user request. |
| |
| The available tools and their respective arguments are: {{ |
| "web_search": ["query"], |
| "visit_webpage": ["url"], |
| }} |
| |
| Return exactly the following format: |
| Response: |
| {{ |
| "tool": "...", |
| "args": {{...}} |
| }} |
| |
| User request: Who nominated the only Featured Article on English Wikipedia about a dinosaur that was promoted in November 2016?. Example of valid JSON expected: |
| Response: |
| {{"tool": "web_search", |
| "args": {{"query": "Who nominated the only Featured Article on English Wikipedia about a dinosaur that was promoted in November 2016?", |
| }} |
| }} |
| |
| Return only one Response! |
| |
| User request: |
| {user_input} |
| """ |
|
|
| output = generate(prompt) |
|
|
| state["proposed_action"] = output.split("Response:")[-1] |
| fixed = repair_json(state["proposed_action"]) |
| data = json.loads(fixed) |
| state["proposed_action"] = data |
|
|
| return state |
|
|
| def safety_node(state: AgentState): |
| """ |
| Evaluate the information provided and output the response for the user request. |
| """ |
|
|
| user_input = state["messages"][-1].content |
| information = state["information"] |
|
|
| prompt = f""" |
| You are a response agent. |
| |
| You must reason over the user request and the provided information and output the answer to the user's request. |
| |
| You MUST return EXACTLY one line in the following format: |
| Response: <answer> |
| |
| DO NOT invent anything additional and return only what is asked and in the format asked. |
| |
| Only return a response if you are confident about the answer, otherwise return empty string. |
| |
| Example of valid json response for user request: Who was the winner of 2025 World Snooker Championship: |
| Response: Zhao Xintong. |
| |
| Return exactly the above requested format and nothing more! |
| DO NOT generate any additional text after it! |
| |
| User request: |
| {user_input} |
| |
| Information: |
| {information} |
| """ |
|
|
| |
| raw_output = generate(prompt) |
|
|
| logger.info(f"Raw Output: {raw_output}") |
|
|
| output = raw_output.split("Response:")[-1].strip() |
| |
| |
|
|
| if len(output) > 2 and output[0] == '"' and output[-1] == '"': |
| output = output[1:-1] |
|
|
| if len(output) > 2 and output[-1] == '.': |
| output = output[:-1] |
|
|
| state["output"] = output |
|
|
| logger.info(f"State (Safety Agent): {state}") |
|
|
| return state |
|
|
|
|
| def Judge(state: AgentState): |
| """ |
| Evaluate whether the answer provided is indeed based on the information provided or not. |
| """ |
|
|
| answer = state["output"] |
| information = state["information"] |
| user_input = state["messages"][-1].content |
|
|
| prompt = f""" |
| You are a Judging agent. |
| |
| You must reason over the user request and judge with a confidence score whether the answer is indeed based on the provided information or not. |
| |
| Example: User request: Who was the winner of 2025 World Snooker Championship? |
| Information: Zhao Xintong won the 2025 World Snooker Championship with a dominant 18-12 final victory over Mark Williams in Sheffield on Monday. The 28 year-old becomes the first player from China to win snooker’s premier prize at the Crucible Theatre. |
| Zhao, who collects a top prize worth £500,000, additionally becomes the first player under amateur status to go all the way to victory in a World Snooker Championship. |
| The former UK champion entered the competition in the very first qualifying round at the English Institute of Sport last month. |
| He compiled a dozen century breaks as he fought his way through four preliminary rounds in fantastic fashion to qualify for the Crucible for the third time in his career. |
| In the final round of the qualifiers known as Judgement Day, Zhao edged Elliot Slessor 10-8 in a high-quality affair during which both players made a hat-trick of tons. |
| Ironically, that probably represented his sternest test throughout the entire event. |
| Answer: "Zhao Xintong" |
| |
| Response: {{ |
| "confidence": 1.0, |
| "explanation": Based on the information provided, it is indeed mentioned that Zhao Xingong, which is the answer provided, won the 2025 World Snooker Championship. |
| }} |
| |
| |
| Example: User request: Who was the winner of 2025 World Snooker Championship? |
| Information: Zhao Xintong won the 2025 World Snooker Championship with a dominant 18-12 final victory over Mark Williams in Sheffield on Monday. The 28 year-old becomes the first player from China to win snooker’s premier prize at the Crucible Theatre. |
| Zhao, who collects a top prize worth £500,000, additionally becomes the first player under amateur status to go all the way to victory in a World Snooker Championship. |
| The former UK champion entered the competition in the very first qualifying round at the English Institute of Sport last month. |
| He compiled a dozen century breaks as he fought his way through four preliminary rounds in fantastic fashion to qualify for the Crucible for the third time in his career. |
| In the final round of the qualifiers known as Judgement Day, Zhao edged Elliot Slessor 10-8 in a high-quality affair during which both players made a hat-trick of tons. |
| Ironically, that probably represented his sternest test throughout the entire event. |
| Answer: "Ronnie O'sullivan" |
| |
| Response: {{ |
| "confidence": 0.0, |
| "explanation": Based on the information provided, it is was Zhao Xingong and not Ronnie O'sullivan who won the 2025 World Snooker Championship. |
| }} |
| |
| Example: User request: Who was the winner of 2025 World Snooker Championship? |
| Information: |
| Answer: "Ronnie O'sullivan" |
| |
| Response: {{ |
| "confidence": 0.0, |
| "explanation": There is no information provided, so cannot answer who won the 2025 World Snooker Championship. |
| }} |
| |
| Return exactly the above requested format and nothing more! |
| DO NOT generate any additional text after it! |
| Return only what is asked and in the format asked! |
| |
| User request: |
| {user_input} |
| |
| Information: |
| {information} |
| |
| Answer: |
| {answer} |
| """ |
|
|
| raw_output = generate(prompt) |
|
|
| print(f"Judge raw output: {raw_output}") |
|
|
| output = raw_output.split("Response:")[-1].strip() |
| fixed = repair_json(output) |
| data = json.loads(fixed) |
|
|
| state["confidence"] = data["confidence"] |
| state["judge_explanation"] = data["explanation"] |
|
|
| logger.info(f"State (Judge Agent): {state}") |
|
|
| return state |
|
|
|
|
| def route(state: AgentState): |
| """Determine the next step based on Safety Agent classification""" |
| if state["risk_score"] > 0.5: |
| return "block" |
| else: |
| return "allow" |
|
|
| def tool_executor(state: AgentState): |
| """ |
| Tool execution node for a risk-aware LLM agent. |
| |
| This node executes the validated and approved tool call proposed by the |
| planner and assessed by the safety layer. It conditionally dispatches |
| execution based on the safety decision and updates the agent state with |
| the final output. |
| |
| Parameters |
| ---------- |
| state : dict |
| Agent state dictionary containing: |
| - "decision" (str): Safety decision ("allow" or blocking variant). |
| - "risk_score" (float): Computed risk score. |
| - "proposed_action" (dict): Validated tool call in structured form. |
| |
| Returns |
| ------- |
| dict |
| Updated state dictionary including: |
| - "output" (str): Result of tool execution OR block message. |
| |
| Execution Flow |
| -------------- |
| 1. If the safety decision is not "allow": |
| - Skip tool execution. |
| - Return a blocked message including the risk score. |
| |
| 2. If allowed: |
| - Validate the proposed action using the `Action` schema. |
| - Dispatch execution to the appropriate tool implementation: |
| * "google_calendar" |
| * "reply_email" |
| * "share_credentials" |
| - Store tool result in `state["output"]`. |
| |
| 3. If the tool is unrecognized: |
| - Return "Unknown tool" as a fallback response. |
| |
| Security Considerations |
| ----------------------- |
| - Execution only occurs after passing the safety node. |
| - No runtime sandboxing is implemented. |
| - No per-tool authorization layer (RBAC) is enforced. |
| - Sensitive tools (e.g., credential exposure) should require: |
| * Elevated approval thresholds |
| * Human-in-the-loop confirmation |
| * Additional auditing |
| |
| Architectural Role |
| ------------------ |
| Planner → Safety → Tool Execution → Logger |
| |
| This node represents the controlled execution layer of the agent, |
| responsible for translating structured LLM intent into real system actions. |
| """ |
|
|
| web_page_result = "" |
| action = Action.model_validate(state["proposed_action"]) |
|
|
| best_query_webpage_information_similarity_score = -1.0 |
| best_webpage_information = "" |
|
|
| webpage_information_complete = "" |
|
|
| if action.tool == "web_search": |
| logger.info(f"action.tool: {action.tool}") |
| |
| query_embeddings = sentence_transformer_model.encode_query(state["messages"][-1].content).reshape(1, -1) |
| query_arg_embeddings = sentence_transformer_model.encode_query(state["proposed_action"]["args"]["query"]).reshape(1, -1) |
| score = float(cosine_similarity(query_embeddings, query_arg_embeddings)[0][0]) |
|
|
| if score > 0.80: |
| results = web_search(**action.args) |
| else: |
| logger.info(f"Overwriting user query because the Agent suggested query had score: {state["proposed_action"]["args"]["query"]} - {score}") |
| results = web_search(**{"query": state["messages"][-1].content}) |
|
|
| logger.info(f"Webpages - Results: {results}") |
|
|
| for result in results: |
| try: |
| web_page_results = visit_webpage(result) |
|
|
| for web_page_result in web_page_results: |
| query_embeddings = sentence_transformer_model.encode_query(state["messages"][-1].content).reshape(1, -1) |
| webpage_information_embeddings = sentence_transformer_model.encode_query(web_page_result).reshape(1, -1) |
| query_webpage_information_similarity_score = float(cosine_similarity(query_embeddings, webpage_information_embeddings)[0][0]) |
| |
| |
| |
| if query_webpage_information_similarity_score > 0.60: |
| webpage_information_complete += web_page_result |
| webpage_information_complete += " \n " |
| webpage_information_complete += " \n " |
| |
| if query_webpage_information_similarity_score > best_query_webpage_information_similarity_score: |
| best_query_webpage_information_similarity_score = query_webpage_information_similarity_score |
| best_webpage_information = web_page_result |
|
|
| except Exception as e: |
| logger.info(f"Tool Executor - Exception: {e}") |
| |
| elif action.tool == "visit_webpage": |
| try: |
| web_page_result = visit_webpage(**action.args) |
| except: |
| pass |
| else: |
| result = "Unknown tool" |
|
|
| state["information"] = webpage_information_complete |
| state["best_query_webpage_information_similarity_score"] = best_query_webpage_information_similarity_score |
|
|
| logger.info(f"Information: {state['information']}") |
| logger.info(f"Information: {state['best_query_webpage_information_similarity_score']}") |
|
|
| return state |
|
|
|
|
| safe_workflow = StateGraph(AgentState) |
| |
|
|
| safe_workflow.add_node("planner", planner_node) |
| safe_workflow.add_node("tool_executor", tool_executor) |
| safe_workflow.add_node("safety", safety_node) |
| |
|
|
| |
|
|
| safe_workflow.add_edge(START, "planner") |
| safe_workflow.add_edge("planner", "tool_executor") |
| safe_workflow.add_edge("tool_executor", "safety") |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
|
|
| |
|
|
|
|
|
|
| |
| |
| class BasicAgent: |
| def __init__(self): |
| self.safe_app = safe_workflow.compile() |
|
|
| 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." |
| |
|
|
|
|
| |
| if " image " not in question and " video " not in question: |
| |
| state = { |
| "messages": question, |
| } |
|
|
|
|
| try: |
| response = self.safe_app.invoke(state) |
| agent_answer = response["output"] |
| except: |
| agent_answer = "" |
| |
| else: |
| agent_answer = fixed_answer |
| |
|
|
|
|
| |
|
|
| return agent_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. |
| """ |
| |
| space_id = os.getenv("SPACE_ID") |
|
|
| 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" |
|
|
| |
| try: |
| agent = BasicAgent() |
| except Exception as e: |
| print(f"Error instantiating agent: {e}") |
| return f"Error initializing agent: {e}", None |
| |
| agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main" |
| print(agent_code) |
|
|
| |
| print(f"Fetching questions from: {questions_url}") |
| try: |
| |
| |
| |
|
|
|
|
| questions_data = [{"task_id":"8e867cd7-cff9-4e6c-867a-ff5ddc2550be","question":"How many studio albums were published by Mercedes Sosa between 2000 and 2009 (included)? You can use the latest 2022 version of english wikipedia.","Level":"1","file_name":""},{"task_id":"a1e91b78-d3d8-4675-bb8d-62741b4b68a6","question":"In the video https://www.youtube.com/watch?v=L1vXCYZAYYM, what is the highest number of bird species to be on camera simultaneously?","Level":"1","file_name":""},{"task_id":"2d83110e-a098-4ebb-9987-066c06fa42d0","question":".rewsna eht sa \"tfel\" drow eht fo etisoppo eht etirw ,ecnetnes siht dnatsrednu uoy fI","Level":"1","file_name":""},{"task_id":"cca530fc-4052-43b2-b130-b30968d8aa44","question":"Review the chess position provided in the image. It is black's turn. Provide the correct next move for black which guarantees a win. Please provide your response in algebraic notation.","Level":"1","file_name":"cca530fc-4052-43b2-b130-b30968d8aa44.png"},{"task_id":"4fc2f1ae-8625-45b5-ab34-ad4433bc21f8","question":"Who nominated the only Featured Article on English Wikipedia about a dinosaur that was promoted in November 2016?","Level":"1","file_name":""},{"task_id":"6f37996b-2ac7-44b0-8e68-6d28256631b4","question":"Given this table defining * on the set S = {a, b, c, d, e}\n\n|*|a|b|c|d|e|\n|---|---|---|---|---|---|\n|a|a|b|c|b|d|\n|b|b|c|a|e|c|\n|c|c|a|b|b|a|\n|d|b|e|b|e|d|\n|e|d|b|a|d|c|\n\nprovide the subset of S involved in any possible counter-examples that prove * is not commutative. Provide your answer as a comma separated list of the elements in the set in alphabetical order.","Level":"1","file_name":""},{"task_id":"9d191bce-651d-4746-be2d-7ef8ecadb9c2","question":"Examine the video at https://www.youtube.com/watch?v=1htKBjuUWec.\n\nWhat does Teal'c say in response to the question \"Isn't that hot?\"","Level":"1","file_name":""},{"task_id":"cabe07ed-9eca-40ea-8ead-410ef5e83f91","question":"What is the surname of the equine veterinarian mentioned in 1.E Exercises from the chemistry materials licensed by Marisa Alviar-Agnew & Henry Agnew under the CK-12 license in LibreText's Introductory Chemistry materials as compiled 08/21/2023?","Level":"1","file_name":""},{"task_id":"3cef3a44-215e-4aed-8e3b-b1e3f08063b7","question":"I'm making a grocery list for my mom, but she's a professor of botany and she's a real stickler when it comes to categorizing things. I need to add different foods to different categories on the grocery list, but if I make a mistake, she won't buy anything inserted in the wrong category. Here's the list I have so far:\n\nmilk, eggs, flour, whole bean coffee, Oreos, sweet potatoes, fresh basil, plums, green beans, rice, corn, bell pepper, whole allspice, acorns, broccoli, celery, zucchini, lettuce, peanuts\n\nI need to make headings for the fruits and vegetables. Could you please create a list of just the vegetables from my list? If you could do that, then I can figure out how to categorize the rest of the list into the appropriate categories. But remember that my mom is a real stickler, so make sure that no botanical fruits end up on the vegetable list, or she won't get them when she's at the store. Please alphabetize the list of vegetables, and place each item in a comma separated list.","Level":"1","file_name":""},{"task_id":"99c9cc74-fdc8-46c6-8f8d-3ce2d3bfeea3","question":"Hi, I'm making a pie but I could use some help with my shopping list. I have everything I need for the crust, but I'm not sure about the filling. I got the recipe from my friend Aditi, but she left it as a voice memo and the speaker on my phone is buzzing so I can't quite make out what she's saying. Could you please listen to the recipe and list all of the ingredients that my friend described? I only want the ingredients for the filling, as I have everything I need to make my favorite pie crust. I've attached the recipe as Strawberry pie.mp3.\n\nIn your response, please only list the ingredients, not any measurements. So if the recipe calls for \"a pinch of salt\" or \"two cups of ripe strawberries\" the ingredients on the list would be \"salt\" and \"ripe strawberries\".\n\nPlease format your response as a comma separated list of ingredients. Also, please alphabetize the ingredients.","Level":"1","file_name":"99c9cc74-fdc8-46c6-8f8d-3ce2d3bfeea3.mp3"},{"task_id":"305ac316-eef6-4446-960a-92d80d542f82","question":"Who did the actor who played Ray in the Polish-language version of Everybody Loves Raymond play in Magda M.? Give only the first name.","Level":"1","file_name":""},{"task_id":"f918266a-b3e0-4914-865d-4faa564f1aef","question":"What is the final numeric output from the attached Python code?","Level":"1","file_name":"f918266a-b3e0-4914-865d-4faa564f1aef.py"},{"task_id":"3f57289b-8c60-48be-bd80-01f8099ca449","question":"How many at bats did the Yankee with the most walks in the 1977 regular season have that same season?","Level":"1","file_name":""},{"task_id":"1f975693-876d-457b-a649-393859e79bf3","question":"Hi, I was out sick from my classes on Friday, so I'm trying to figure out what I need to study for my Calculus mid-term next week. My friend from class sent me an audio recording of Professor Willowbrook giving out the recommended reading for the test, but my headphones are broken :(\n\nCould you please listen to the recording for me and tell me the page numbers I'm supposed to go over? I've attached a file called Homework.mp3 that has the recording. Please provide just the page numbers as a comma-delimited list. And please provide the list in ascending order.","Level":"1","file_name":"1f975693-876d-457b-a649-393859e79bf3.mp3"},{"task_id":"840bfca7-4f7b-481a-8794-c560c340185d","question":"On June 6, 2023, an article by Carolyn Collins Petersen was published in Universe Today. This article mentions a team that produced a paper about their observations, linked at the bottom of the article. Find this paper. Under what NASA award number was the work performed by R. G. Arendt supported by?","Level":"1","file_name":""},{"task_id":"bda648d7-d618-4883-88f4-3466eabd860e","question":"Where were the Vietnamese specimens described by Kuznetzov in Nedoshivina's 2010 paper eventually deposited? Just give me the city name without abbreviations.","Level":"1","file_name":""},{"task_id":"cf106601-ab4f-4af9-b045-5295fe67b37d","question":"What country had the least number of athletes at the 1928 Summer Olympics? If there's a tie for a number of athletes, return the first in alphabetical order. Give the IOC country code as your answer.","Level":"1","file_name":""},{"task_id":"a0c07678-e491-4bbc-8f0b-07405144218f","question":"Who are the pitchers with the number before and after Taishō Tamai's number as of July 2023? Give them to me in the form Pitcher Before, Pitcher After, use their last names only, in Roman characters.","Level":"1","file_name":""},{"task_id":"7bd855d8-463d-4ed5-93ca-5fe35145f733","question":"The attached Excel file contains the sales of menu items for a local fast-food chain. What were the total sales that the chain made from food (not including drinks)? Express your answer in USD with two decimal places.","Level":"1","file_name":"7bd855d8-463d-4ed5-93ca-5fe35145f733.xlsx"},{"task_id":"5a0c1adf-205e-4841-a666-7c3ef95def9d","question":"What is the first name of the only Malko Competition recipient from the 20th Century (after 1977) whose nationality on record is a country that no longer exists?","Level":"1","file_name":""}] |
|
|
|
|
| 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 |
|
|
|
|
|
|
|
|
|
|
| |
| |
| |
| |
| |
| |
|
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
|
|
| |
| |
| |
|
|
| |
| |
| |
| |
| |
| |
|
|
|
|
|
|
|
|
| |
|
|
| |
| 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) |
|
|
| |
| 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) |
|
|
| |
| 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 |
|
|
|
|
| |
| 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) |
| |
| 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") |
|
|
| 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(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) |
