--- title: "Python Notebook Sample Builder: review, role & definition · Undominated.ai" canonical: https://undominated.ai/agents/github-python-notebook-sample-builder/ description: "Builds a new Python demonstration notebook from snippets checked in the selected development environment." --- # Python Notebook Sample Builder: review, role & definition · Undominated.ai > Builds a new Python demonstration notebook from snippets checked in the selected development environment. [← Explore all agents](/agents/) DATA AND ANALYTICS / GitHub community # Python Notebook Sample Builder Builds a new Python demonstration notebook from snippets checked in the selected development environment. [Use this definition ↓](#setup)[Original source ↗](https://github.com/github/awesome-copilot/blob/3a685010a7afdc0dbd4c83b7fbda6c316aa516e5/agents/python-notebook-sample-builder.agent.md) SOURCE REVIEW Reviewed 2026-10-14 Evidence 3 linked sources Publisher GitHub community Licence [MIT ↗](https://github.com/github/awesome-copilot/blob/3a685010a7afdc0dbd4c83b7fbda6c316aa516e5/LICENSE) Revision 3a685010a7af [Read what was—and wasn’t—checked ↓](#review) “Custom agent for building Python Notebooks in VS Code that demonstrate Azure and AI features” [GitHub community · upstream description ↗](https://github.com/github/awesome-copilot/blob/3a685010a7afdc0dbd4c83b7fbda6c316aa516e5/agents/python-notebook-sample-builder.agent.md) Our analysis follows below. 01 / THE REASONING ## Why this made the selection. - The file says not to put code in a notebook until it has been run in the terminal. - It also forbids internal-only APIs and says every sample must use public SDKs. ### A good fit for - Supply the demonstration goal, existing notebook style, configured devcontainer/kernel and authorized public Azure/AI services. - Receive a new notebook assembled from terminal-tested snippets with prerequisites, concise sections and visual outputs. ### Weigh up before choosing - The prompt assumes a development container and broad package-install permissions; do not apply that assumption to an arbitrary host. - Python, Jupyter and Microsoft Learn tool identifiers depend on installed VS Code integrations. - Executing notebook cells can call paid APIs or change external data. Use authorized test inputs and inspect those effects before running. - Terminal execution alone does not prove clean-kernel notebook execution order; rerun the assembled notebook from a fresh kernel. 02 / THE REVIEW RECORD ## What we actually inspected. Source review has boundaries. A clear record is more useful than a “safe” badge. ### Material inspected - agents/python-notebook-sample-builder.agent.md - LICENSE - Definition reread at the preserved revision; scoped workflow and output-contract review on 2026-10-14 ### Our findings - Frontmatter tools include mslearnmcp/*, Jupyter notebook configuration, and installNotebookPackages. - The file says never to include notebook code that has not been run and verified in the terminal. - It says all code must work with publicly available SDKs and documentation. - Research says to use Microsoft Learn and then validate against the actual SDK locally. - The prompt explicitly requires a new file rather than overwriting an existing notebook. - It separates Microsoft Learn research from actual SDK execution and requires each snippet to run before it enters a notebook cell. ### Not established by this review - The agent was not executed against a live repository or host. The review applies to the material and revision named here. A newer upstream release can change its behavior. 03 / PUT IT TO WORK ## Use the role in your project. [Upstream setup instructions ↗](https://code.visualstudio.com/docs/copilot/customization/custom-agents) - Download the original definition with its licence and attribution; inspect the prompt before loading it. - Put a reviewed working copy at .github/agents/python-notebook-sample-builder.agent.md in the intended project. - Check the model and tool identifiers supported by your client; grant only the tools needed for the selected task. - Select the agent in your client and begin with a bounded task whose result you can independently inspect. ### Before you start - VS Code GitHub Copilot with custom-agent support - An explicitly prepared Python development container - The required VS Code Python/Jupyter integrations and any optional Microsoft Learn connector ### Compatibility VS Code GitHub Copilot ### Agent instructions and host-provided tools - vscode, execute, read, edit, search, web, mslearnmcp/*, Python environment tools, and Jupyter package tools. ### Cost model MIT. The file does not state a hosted-model price. THE COMPLETE REVIEWED DEFINITION ## Read it before you reuse it. Original source bytes, with attribution. Review the host-specific setup notes above. Copy definition ↗ [Download definition + licence ↗](/resources/agents/github-python-notebook-sample-builder/bundle.zip)[Raw Markdown ↗](/resources/agents/github-python-notebook-sample-builder/definition.md) --- description: 'Custom agent for building Python Notebooks in VS Code that demonstrate Azure and AI features' name: 'Python Notebook Sample Builder' tools: ['vscode', 'execute', 'read', 'edit', 'search', 'web', 'mslearnmcp/*', 'agent', 'ms-python.python/getPythonEnvironmentInfo', 'ms-python.python/getPythonExecutableCommand', 'ms-python.python/installPythonPackage', 'ms-python.python/configurePythonEnvironment', 'ms-toolsai.jupyter/configureNotebook', 'ms-toolsai.jupyter/listNotebookPackages', 'ms-toolsai.jupyter/installNotebookPackages', 'todo'] --- You are a Python Notebook Sample Builder. Your goal is to create polished, interactive Python notebooks that demonstrate Azure and AI features through hands-on learning. ## Core Principles - **Test before you write.** Never include code in a notebook that you have not run and verified in the terminal first. If something errors, troubleshoot the SDK or API until you understand the correct usage. - **Learn by doing.** Notebooks should be interactive and engaging. Minimize walls of text. Prefer short, crisp markdown cells that set up the next code cell. - **Visualize everything.** Use built-in notebook visualization (tables, rich output) and common data science libraries (matplotlib, pandas, seaborn) to make results tangible. - **No internal tooling.** Avoid any internal-only APIs, endpoints, packages, or configurations. All code must work with publicly available SDKs, services, and documentation. - **No virtual environments.** We are working inside a devcontainer. Install packages directly. ## Workflow 1. **Understand the ask.** Read what the user wants demonstrated. The user's description is the master context. 2. **Research.** Use Microsoft Learn to investigate correct API usage and find code samples. Documentation may be outdated, so always validate against the actual SDK by running code locally first. 3. **Match existing style.** If the repository already contains similar notebooks, imitate their structure, style, and depth. 4. **Prototype in the terminal.** Run every code snippet before placing it in a notebook cell. Fix errors immediately. 5. **Build the notebook.** Assemble verified code into a well-structured notebook with: - A title and brief intro (markdown) - Prerequisites / setup cell (installs, imports) - Logical sections that build on each other - Visualizations and formatted output - A summary or next-steps cell at the end 6. **Create a new file.** Always create a new notebook file rather than overwriting existing ones. ## Notebook Structure Guidelines - **Title cell** — One `#` heading with a concise title. One sentence describing what the reader will learn. - **Setup cell** — Install dependencies (`%pip install ...`) and import libraries. - **Section cells** — Each section has a short markdown intro followed by one or more code cells. Keep markdown crisp: 2-3 sentences max per cell. - **Visualization cells** — Use pandas DataFrames for tabular data, matplotlib/seaborn for charts. Add titles and labels. - **Wrap-up cell** — Summarize what was covered and suggest next steps or further reading. ## Style Rules - Use clear variable names and inline comments where the intent is not obvious. - Prefer f-strings for string formatting. - Keep code cells focused: one concept per cell. - Use `display()` or rich DataFrame rendering instead of plain `print()` for tabular data. - Add `# Section Title` comments at the top of code cells for scanability. The download contains python-notebook-sample-builder.agent.md . Keep its filename when placing it in the agent directory described above. By **GitHub community**. [Exact upstream source ↗](https://raw.githubusercontent.com/github/awesome-copilot/3a685010a7afdc0dbd4c83b7fbda6c316aa516e5/agents/python-notebook-sample-builder.agent.md) · [Licence](/resources/agents/github-python-notebook-sample-builder/LICENSE.txt) · [Attribution](/resources/agents/github-python-notebook-sample-builder/ATTRIBUTION.txt) SHA-256 a2dd06852ba4655565f9daecd939fb7ee87f7bb789442a555a706df3ed7e7980 Read the applicable licence MIT License Copyright GitHub, Inc. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. 04 / FOLLOW THE EVIDENCE ## The source trail. Our notes are separate from the original resource. Check upstream before adopting a new version. - [Definition at the reviewed revision ↗](https://github.com/github/awesome-copilot/blob/3a685010a7afdc0dbd4c83b7fbda6c316aa516e5/agents/python-notebook-sample-builder.agent.md) Checked 2026-10-14 https://github.com/github/awesome-copilot/blob/3a685010a7afdc0dbd4c83b7fbda6c316aa516e5/agents/python-notebook-sample-builder.agent.md Supports: summary, upstreamDescription, whySelected, bestFor, limitations, review, access - [Licence at the reviewed revision ↗](https://github.com/github/awesome-copilot/blob/3a685010a7afdc0dbd4c83b7fbda6c316aa516e5/LICENSE) Checked 2026-10-14 https://github.com/github/awesome-copilot/blob/3a685010a7afdc0dbd4c83b7fbda6c316aa516e5/LICENSE Supports: license, access.cost - [Host custom-agent files and tool permissions ↗](https://code.visualstudio.com/docs/copilot/customization/custom-agents) Checked 2026-10-07 https://code.visualstudio.com/docs/copilot/customization/custom-agents Supports: install, compatibility, limitations, access KEEP COMPARING ## Other approaches to consider. Related by category or shared topics. These are alternatives to inspect, not a measured quality order. [### Data Engineer ↗ A broad design prompt for batch and streaming pipelines, data quality, schema evolution and operational handoffs.](/agents/wshobson-data-engineer/)[### PySpark Expert Agent ↗ Reviews PySpark pipeline code for distributed-execution anti-patterns with confidence-graded hypotheses, Spark-native fixes, and a ban on inventing runtime evidence.](/agents/github-spark-performance/)[### Spatial Data Engineer ↗ A geospatial ETL planning prompt centered on source preservation, explicit coordinate systems and transformation checks.](/agents/agency-spatial-data-engineer/) [AI Tools ↗](/tools/)[Skills ↗](/skills/)[Agents ↗](/agents/)[MCP Servers ↗](/mcp-servers/)[Workflows ↗](/workflows/) ## Continue your investigation - [Find reusable instructions](/skills/) - [Inspect connections](/mcp-servers/) - [Build a toolkit](/tools/) - [Choose the model](/compare/)