Logseq Skills

A collection of custom AI assistant skills designed for Logseq users. These skills enhance AI assistants (like Claude) to better support Logseq workflows.

Overview

Logseq Skills provides a set of specially optimized prompt templates that enable AI assistants to better understand and utilize Logseq features such as block references, properties, queries, and more. Each skill is an independent module focused on specific use cases.

Included Skills

Core Skills

Skill Name	Description
logseq-outline	Format any content as a Logseq-compatible outline using nested bullets (no Markdown headers)
logseq-inbox-capture	Capture content into a Logseq inbox page with automatic metadata and properties
logseq-inbox-triage	Classify and process items in your inbox
logseq-weekly-review	Conduct weekly reviews to organize and review the past week's content

GTD-Related Skills

Skill Name	Description
logseq-gtd-project-pages	Create and manage GTD-style project pages
logseq-gtd-contexts	Organize tasks using context tags (@computer, @phone, etc.)

Knowledge Management Skills

Skill Name	Description
logseq-file-graph-knowledge	Understand and use Logseq's file graph (Markdown mode) for knowledge management
logseq-dashboard-file-graph	Create dashboards and query views for file graphs

Usage

Using with Claude Desktop

1. Clone or download this repository to your local machine
2. Configure the skills path in Claude Desktop:
   • Open Claude Desktop settings
   • Find the Skills or Custom Skills configuration section
   • Add this repository's path
3. Trigger skills in conversation using natural language keywords

Using with Claude Code

Claude Code Skills are automatically discovered when placed in the correct directories.

Clone the Repository

git clone https://github.com/vlooong/logseq-skills.git
cd logseq-skills

Configure Skills

# Create skills directory in your logseq directory
mkdir -p .claude/skills

# Copy all skill directories from logseq-skills to .claude/skills
cp -r logseq-skills/logseq-* .claude/skills/

These Skills are available to anyone working in this repository. Commit .claude/skills/ to version control to share them with your team.

Use Skills

Skills are triggered automatically based on your conversation. For example:

• "Format this as a Logseq outline"
• "Add this to my inbox"
• "Create a GTD project page"
• "I need to do a weekly review"

Claude will automatically detect the relevant skill and ask permission to use it.

Manual Invocation

You can also manually invoke a skill using the / command:

/logseq-outline

This will load the skill and apply it to your current context.

Skill Trigger Examples

Most skills can be triggered through natural language, for example:

• "Format this as a Logseq outline"
• "Add this to my inbox"
• "Create a GTD project page"
• "I need to do a weekly review"

Skill Structure

Each skill contains a SKILL.md file with the following structure:

---
name: skill-name
description: Brief description of the skill
---

# Skill Title

Detailed description of the skill...

## When to Use
Usage scenarios and trigger phrases

## Core Rules
Core rules and non-negotiable requirements

## Examples
Usage examples

## Quality Checklist
Quality assurance checklist

File Graph vs Database Graph

Skills in this repository are primarily designed for Logseq's file graph mode (Markdown mode), using:

• TODO / DOING / DONE task markers
• key:: value property syntax
• {{query ...}} simple query syntax

They do not use database graph-specific concepts (such as #Task classes, Status property tasks, etc.).

Contributing

Contributions of new skills or improvements to existing skills are welcome!

Adding a New Skill

1. Create a new subfolder in the repository root, named: logseq-{skill-name}
2. Create a SKILL.md file in that folder
3. Add necessary frontmatter:

---
name: logseq-your-skill-name
description: Your skill description, one sentence explaining the main functionality
---

4. Write content following the skill template, which should include:
   • Clear usage scenarios
   • Core rules and constraints
   • Practical examples
   • Quality checklist

Best Practices for Writing Skills

• Keep it Simple: Each skill should focus on a single responsibility
• Practical Focus: Provide content that can be directly copied and pasted
• Rich Examples: Include examples covering multiple use cases
• Quality Assurance: Provide checklists to ensure output quality
• Version Compatibility: Clearly indicate supported Logseq versions and modes (file graph/database graph)

Commit Convention

Commit messages should clearly describe changes:

feat: add new skill logseq-task-calendar
fix: fix formatting issue in logseq-outline
docs: update usage instructions in README

Code Review

All contributions will undergo the following review:

• Compliance with existing skill structure
• Clear examples provided
• Actually tested
• No conflicts with existing skills

License

This project is licensed under the MIT License. See LICENSE file for details.

Related Resources

• Logseq Official Documentation
• Logseq Community
• Claude Code Documentation

Feedback & Suggestions

For questions or suggestions:

• Submit an Issue
• Open a Pull Request
• Discuss in the Logseq community

---

Note: This repository is under active development. Skill APIs and formats may change with version updates.