What are the benefits of the 3-layer architecture over standard prompts?

By separating constraints from the prompt builder, you gain better testability, reusability of validation rules, and more consistent AI responses compared to monolithic prompts.

Which LLMs are supported for field suggestions?

The provided examples use Ollama with Qwen 2.5, but the 3-layer architecture is designed to be model-agnostic and can be adapted for OpenAI, Anthropic, or other providers.

Is this compatible with the Dataview plugin?

Yes, the architecture specifically includes constraints that enforce wikilink formatting and specific enum values required for clean Dataview queries.

How does this skill interact with my existing Obsidian vault?

The skill uses a 'Vault Context' object to ingest your current areas, resources, and tags, allowing the AI to suggest existing links rather than creating redundant new ones.

PARA Obsidian Field Suggestions

Name: PARA Obsidian Field Suggestions
Author: nathanvale

bynathanvale

•

Productivity & Workflow

Automates metadata extraction and frontmatter suggestions for Obsidian vaults using a structured 3-layer LLM architecture.

About

The Field Suggestions skill provides a robust framework for integrating AI-assisted metadata management into Obsidian workflows, specifically optimized for the PARA method. By leveraging a specialized 3-layer architecture—encompassing deterministic constraints, declarative prompt building, and high-level orchestration—it enables developers to build tools that automatically suggest project areas, tags, and status updates while respecting existing vault context and Dataview formatting rules. This skill is ideal for users who want to maintain a highly organized knowledge base without the cognitive overhead of manual entry, ensuring that every note is correctly categorized and linked within their personal productivity system.

Key Features

Deterministic Validation: Enforces wikilink formatting and enum rules to ensure Dataview compatibility
3-Layer LLM Architecture: Separates constraints, prompt logic, and orchestration for maximum reliability
Local LLM Integration: Optimized for privacy-focused workflows using Ollama and models like Qwen 2.5
Flexible Orchestration: Includes helpers for both single-field suggestions and full note-to-template conversion
Vault-Aware Extraction: Automatically respects existing PARA areas, resources, and tags during suggestions
1 GitHub stars

Use Cases

Building interactive slash commands that provide real-time metadata suggestions during note creation
Automating the migration of legacy notes to a new metadata schema with consistent validation
Converting messy, unstructured scratchpad notes into formatted PARA project files with valid frontmatter

About

Key Features

Deterministic Validation: Enforces wikilink formatting and enum rules to ensure Dataview compatibility
3-Layer LLM Architecture: Separates constraints, prompt logic, and orchestration for maximum reliability
Local LLM Integration: Optimized for privacy-focused workflows using Ollama and models like Qwen 2.5
Flexible Orchestration: Includes helpers for both single-field suggestions and full note-to-template conversion
Vault-Aware Extraction: Automatically respects existing PARA areas, resources, and tags during suggestions
1 GitHub stars

Use Cases

Building interactive slash commands that provide real-time metadata suggestions during note creation
Automating the migration of legacy notes to a new metadata schema with consistent validation
Converting messy, unstructured scratchpad notes into formatted PARA project files with valid frontmatter