Conversational AI Agent Overview
On this page
Conversational AI Agent Overview
The VibeMap Conversational Agent is a chat-based AI assistant built directly into your project editor. It understands your entire project -- every persona, feature, user story, acceptance criterion, and schema entity -- and lets you refine your specification through natural conversation.
What Is the Conversational Agent?
The agent is more than a chatbot. It is a context-aware project co-pilot that reads your project's current state before every response. When you ask it to add a feature or modify a persona, it already knows what exists, how entities relate to each other, and what the downstream effects of a change might be.
You interact with it through the chat panel on the right side of the project editor.
What Can It Do?
The agent supports a wide range of project operations:
| Capability | Examples |
|---|---|
| Add entities | Create new features, personas, user stories, acceptance criteria |
| Edit entities | Rename a feature, update a persona's goals, refine story details |
| Remove entities | Delete outdated personas or features (with impact warnings) |
| Answer questions | "How many features reference the admin persona?" |
| Analyze relationships | "What user stories depend on the authentication feature?" |
| Bulk operations | "Add acceptance criteria to all features that don't have any" |
| Suggest improvements | "What's missing from my project?" |
| Run generations | One-click suggested actions that run the same pipelines as the page buttons |
Suggested Actions
The agent panel shows contextual suggested action chips — in the empty state and above the message input. They come in two kinds:
- Pipeline chips (wand/sparkles icon) run a full generation — personas, features, user stories, criteria derivation, pages, schema, or the business case — through exactly the same pipeline as the button on that section's page. Billing, live progress, and version control behave identically. The highlighted chip is the recommended next step for your project's current state.
- Prefill chips (pen icon) drop an editable question into the input, such as "Find gaps in my spec" — nothing runs until you press send.
The chips are derived from what already exists in your project: a project with features but no user stories will suggest generating stories; once pages and criteria both exist, you'll see a "Sync criteria from pages" chip that cross-checks acceptance criteria against your page layouts. Chip-triggered generations use whichever model you've selected in the input's model picker.
How It Understands Your Project
Every time you send a message, the agent goes through a structured process:
- Intent classification -- Determines whether you are asking a question, requesting a generation, or modifying an existing entity.
- Context retrieval -- Pulls the relevant parts of your project (features, personas, stories, schema) into its working memory.
- Impact analysis -- For modifications, it checks how the change would affect related entities and warns you of any cascading effects.
- Execution -- Performs the operation and updates your project in real time.
This means the agent always responds with awareness of what you have already built, keeping new additions consistent with your existing specification.
When to Use the Agent vs. Regeneration
Both the agent and the regeneration buttons can create and update project content, but they serve different purposes:
Use the Agent When You Want To:
- Make a specific, targeted change ("Rename the Notifications feature to Alerts")
- Add a single entity with particular requirements ("Add a persona for a compliance officer")
- Ask questions about your project's current state
- Perform multi-step operations in one request
- Understand how a change would affect other parts of your project
Use Regeneration When You Want To:
- Generate a full set of entities from scratch (e.g., all personas at once)
- Replace an entire section with fresh AI-generated content
- Start over on a particular generation step
- Produce content in bulk without specific instructions
A Simple Rule of Thumb
If you know exactly what you want changed, use the agent. If you want the AI to make broad decisions for you, use regeneration.
Key Benefits
- Natural language control -- No forms or dropdowns. Just describe what you need in plain English.
- Contextual accuracy -- The agent references your existing project data, so new content fits naturally.
- Safety guardrails -- Destructive operations (like deletions) require confirmation and show impact analysis before proceeding.
- Real-time updates -- Changes appear immediately in your project editor. No page refresh needed.
- Conversation history -- The agent remembers your earlier messages within the session, so you can build on previous requests.