| name | query-project-info |
| description | Project metadata extraction skill for Revit Snapshots. Retrieves global model attributes (Name, Number, Client, Address) and model inventory (Levels, Sheets, Links). |
Skill: Project Metadata & Inventory
This skill provides a standardized set of queries for auditing top-level project information and model structure. It acts as the "Project Passport," giving agents immediate context about the model's identity and discipline.
1. Core Logic & Assumptions
Categorical Filtering
- Metadata: Extracted from the
elements table where category_name = 'Project Information'.
- Inventory: Levels, Sheets, and Links are identified by their specific
category_name (e.g., Levels, Sheets).
Scoping
This skill distinguishes between the active document (scope: 'active') and the entire loaded ecosystem (scope: 'global').
2. Capability Manifest (Query Map)
The skill is modularized into dedicated SQL files:
| Query File | Description | Output |
|---|
extract-project-info.sql | Project Passport | Name, Number, Client, Discipline, Address. |
list-levels.sql | Spatial Context | Elevation-sorted list of project levels. |
list-sheets.sql | Drawing List | Comprehensive sheet index. |
list-linked-models.sql | Link Registry | Inventory of all attached Revit models. |
3. Best Practices for Developers
- Verification First: Always run
extract-project-info.sql during the planning phase to confirm you are in the correct model.
- Elevation Logic: When listing levels, always sort by
elevation to maintain spatial hierarchy.
- Global Awareness: Use the
global/ variant of these queries if the task requires cross-model coordination (e.g., "Find all architectural sheets across all links").
4. When to use this skill
- At the start of a task to gather project-wide context.
- When the user asks for high-level statistics (e.g., "How many levels are in this building?").
- When auditing sheet numbers or project addresses for documentation.
5. When NOT to use this skill
- For quantitative element takeoffs (Pipes, Walls). Use category-specific analytical skills instead.
- For detailed parameter inspection of individual instances.