microsoft/hve-core
Publicmirrored fromhttps://github.com/microsoft/hve-coreAvailable
docs/getting-started/first-interaction.md
75lines · modecode
| 1 | --- |
| 2 | title: Your First Interaction |
| 3 | description: Talk to an HVE Core agent and see it respond in under 60 seconds |
| 4 | author: Microsoft |
| 5 | ms.date: 2026-02-19 |
| 6 | ms.topic: tutorial |
| 7 | keywords: |
| 8 | - getting started |
| 9 | - first interaction |
| 10 | - memory agent |
| 11 | - github copilot |
| 12 | estimated_reading_time: 2 |
| 13 | --- |
| 14 | |
| 15 | > [!NOTE] |
| 16 | > Step 1 of 4 in the [Getting Started Journey](README.md). |
| 17 | |
| 18 | Before diving into workflows and methodologies, confirm that everything works. |
| 19 | You need one agent interaction that produces a visible result. |
| 20 | |
| 21 | ## Talk to the Memory Agent |
| 22 | |
| 23 | 1. Open GitHub Copilot Chat (`Ctrl+Alt+I`). |
| 24 | 2. Select the **memory** agent from the agent picker. |
| 25 | 3. Type this prompt: |
| 26 | |
| 27 | > Remember that I am a [your role] and I'm learning HVE Core for the first |
| 28 | > time. |
| 29 | |
| 30 | Replace `[your role]` with your actual role, such as *software engineer*, |
| 31 | *tech lead*, *product manager*, or *designer*. |
| 32 | |
| 33 | 4. The agent responds and creates a file in your workspace under `memories/`. |
| 34 | Open it. You'll see your role stored as a note that persists across |
| 35 | sessions. |
| 36 | |
| 37 | ## See Memory in Action |
| 38 | |
| 39 | Now verify that other agents can read what the memory agent stored. |
| 40 | |
| 41 | 1. Open a new Copilot Chat thread. |
| 42 | 2. Type this prompt: |
| 43 | |
| 44 | > Explain what this repository does and how it helps someone in my role. |
| 45 | |
| 46 | 3. The response references your role without you mentioning it again. Copilot |
| 47 | read the memory file, found your stored context, and tailored the |
| 48 | explanation. |
| 49 | |
| 50 | You just proved four things: HVE Core is installed, agents respond to natural |
| 51 | language, the memory system creates real files, and other agents use those |
| 52 | files to personalize their responses. |
| 53 | |
| 54 | ## What Is the Memory Agent? |
| 55 | |
| 56 | The memory agent stores notes that persist across conversations. Agents and |
| 57 | prompts can read these notes to personalize their behavior. When you told it |
| 58 | your role, every future interaction can use that context without you repeating |
| 59 | it. |
| 60 | |
| 61 | This is a small example of a larger pattern in HVE Core: agents produce |
| 62 | artifacts (files, documents, plans) rather than chat responses alone. The |
| 63 | artifacts carry context forward so you don't repeat yourself. |
| 64 | |
| 65 | ## Next Step |
| 66 | |
| 67 | Now that you know agents work, try using one for real work: |
| 68 | [Your First Research](first-research.md). |
| 69 | |
| 70 | --- |
| 71 | |
| 72 | <!-- markdownlint-disable MD036 --> |
| 73 | *🤖 Crafted with precision by ✨Copilot following brilliant human instruction, |
| 74 | then carefully refined by our team of discerning human reviewers.* |
| 75 | <!-- markdownlint-enable MD036 --> |
| 76 | |