Spaces:
Running
Running
| title: Tool-Using Agent | |
| emoji: 🛠️ | |
| colorFrom: indigo | |
| colorTo: purple | |
| sdk: docker | |
| app_port: 7860 | |
| pinned: false | |
| short_description: An AI agent that calls live tools (ReAct) · 30-in-15 #9 | |
| # Tool-Using Agent 🛠️ | |
| Ask a question and watch the AI **call real, live tools** to answer it — a calculator, the current | |
| date/time, live **weather**, and **Wikipedia** — running the **ReAct** loop (decide → call a tool → | |
| read the result → repeat) and showing every tool call it made. | |
| > Project **#9** of my *"30 AI Projects in 15 Days"* build-in-public challenge. | |
| **▶ Live demo: https://agent.gritai.solutions** | |
|  | |
| ## Try | |
| - *"What's a 15% tip on $47.80, and what's the weather in Paris?"* → calculator + weather | |
| - *"Who invented the World Wide Web, and how many days ago was Jan 1 2000?"* → Wikipedia + date + calc | |
| ## How it works | |
| - The model is given four **keyless** tools (Open-Meteo for weather, MediaWiki for Wikipedia, plus a | |
| safe calculator and a clock). It decides which to call — possibly several — and the server executes | |
| them and feeds results back, looping until the model can answer (capped at a few steps). | |
| - **Transparency:** the UI shows the exact tool calls and their live results, so you can see the | |
| reasoning, not just the answer. | |
| - The calculator is a **safe AST evaluator** (no `eval`), limited to arithmetic + a few math functions. | |
| ## Run locally | |
| ```bash | |
| pip install -r requirements.txt | |
| export ANTHROPIC_API_KEY=sk-ant-... | |
| python app.py # http://127.0.0.1:7860 | |
| ``` | |
| ## Deploy (always-on) | |
| Ships with a `Dockerfile` — works on **Hugging Face Spaces**, Render, Railway, or Fly.io. | |
| Set `ANTHROPIC_API_KEY` as a secret. The tools need no other keys. | |
| --- | |
| Built by **Robert Lucyk** · [GritAI Solutions](https://gritai.solutions) · part of the 30-in-15 challenge. | |