AI Advisor
The AI Advisor is an optional feature that reviews your diagrams and suggests improvements. It runs in the background and surfaces hints as small badges on nodes.
What it checks
Section titled “What it checks”The advisor evaluates your model against C4 best practices:
- Relationships — missing, misleading, or wrong-direction edges
- Node naming — technology-stuffed names that should describe a role instead
- Description quality — vague or placeholder descriptions, and prose-shaped descriptions that should be rewritten as verb-led bullets
- Structural issues — client-side apps talking directly to a database, missing queues between async services, unclear boundaries
- Missing infrastructure — no auth, validation, rate limiting, or migrations where production code would need them
- Flow step granularity — steps that describe UI gestures instead of system interactions
How hints appear
Section titled “How hints appear”When the advisor finds something, a colored dot appears on the affected node:
- 🟠 Orange dot — a warning that likely needs attention
- 🟢 Teal dot — informational suggestion
Select the node to see the hint message in the Properties panel. Some hints come with a one-click fix you can apply right from there.
Dismissing hints
Section titled “Dismissing hints”If a hint isn’t relevant, check the dismiss box in the Properties panel. Dismissed hints won’t reappear for that node unless the model changes significantly.
Setting up the advisor
Section titled “Setting up the advisor”Open Settings (gear icon in the top bar) and go to the AI Review tab.
Choose a provider and enter your API key:
| Provider | Models |
|---|---|
| OpenAI | GPT-4o, GPT-4, etc. |
| Anthropic | Claude Sonnet, Claude Opus, etc. |
| Gemini Pro, etc. | |
| Groq | Mixtral, Llama, etc. |
| Mistral | Mistral Large, etc. |
| DeepSeek | DeepSeek Chat, etc. |
| Ollama | Any locally running model |
Click Fetch models to see available models for your provider. Select one and save.
When it runs
Section titled “When it runs”The advisor runs automatically after a 10-second pause following model changes. It doesn’t run on every keystroke — only when the model’s structure changes (adding/removing nodes, changing edges, etc.).
The advisor is purely optional. Scryer works fine without it — it’s just an extra layer of review.
Privacy
Section titled “Privacy”When using cloud providers, your model’s structure (node names, descriptions, edges) is sent to the API. No source code is sent. If you prefer to keep everything local, use Ollama with a locally running model.