linkly-ai-skills
Agent Skills for Linkly AI — search, browse, and read your local documents from any AI coding agent.
Linkly AI Skills
Agent Skills for Linkly AI — search, browse, and read your local documents from any AI coding agent.
This skill teaches AI agents how to use Linkly AI's document search capabilities, enabling them to find and read your locally indexed documents (PDF, Markdown, DOCX, TXT, HTML, and more).
What is Linkly AI?
Linkly AI is a desktop application that indexes documents on your computer and provides full-text search, structural outlines, and content reading through a local MCP server. Think of it as a local knowledge base that AI agents can query.
What Does This Skill Do?
When installed, this skill enables AI agents to:
- Search your local documents by keywords with relevance ranking
- List libraries to discover and search within specific knowledge bases
- Browse document outlines to understand structure before diving in
- Grep for specific text patterns with regex matching
- Read document content with line-based pagination
- Diagnose issues with
linkly doctorwhen things aren't working - Auto-detect whether to use CLI commands or MCP tools based on the environment
- Guide setup if Linkly AI is not yet installed
The skill supports two access modes:
| Mode | When Used | How It Works |
|---|---|---|
| CLI | Agent has Bash/terminal access | Runs linkly CLI commands (preferred) |
| MCP | Agent has MCP tool access | Calls search/outline/grep/read/list_libraries MCP tools |
Prerequisites
- Linkly AI desktop app — download from linkly.ai
- Linkly AI CLI (for CLI mode) — see installation
CLI Installation
macOS / Linux:
curl -sSL https://updater.linkly.ai/cli/install.sh | sh
Or via Homebrew:
brew tap LinklyAI/tap
brew install linkly
Windows (PowerShell):
irm https://updater.linkly.ai/cli/install.ps1 | iex
Cross-platform (requires Rust):
cargo install linkly-ai-cli
Installing This Skill
skills.sh (Recommended)
Install to all supported agents with a single command:
npx skills add LinklyAI/linkly-ai-skills
Or install to a specific agent:
# Claude Code only
npx skills add LinklyAI/linkly-ai-skills -a claude-code
# Codex CLI only
npx skills add LinklyAI/linkly-ai-skills -a codex
# Global install (available across all projects)
npx skills add LinklyAI/linkly-ai-skills -g
Claude Code (manual)
Copy the skill to your personal skills directory:
git clone https://github.com/LinklyAI/linkly-ai-skills.git ~/.claude/skills/linkly-ai
Or for a specific project:
git clone https://github.com/LinklyAI/linkly-ai-skills.git .claude/skills/linkly-ai
Codex CLI (OpenAI)
git clone https://github.com/LinklyAI/linkly-ai-skills.git ~/.agents/skills/linkly-ai
Claude.ai (web)
Download linkly-ai.zip from the Releases page, then upload it in Claude.ai → Settings → Capabilities → Skills.
ClawHub (OpenClaw)
clawhub install linkly-ai
Other AI Agents
Any AI agent that supports the Agent Skills open standard can use this skill. Copy the SKILL.md file and the references/ directory to the appropriate skills location for your agent.
Skill Contents
├── SKILL.md # Core skill instructions
├── references/
│ ├── cli-reference.md # CLI commands and options
│ ├── mcp-tools-reference.md # MCP tool schemas and responses
│ ├── search-strategies.md # Advanced query crafting patterns
│ └── troubleshooting.md # Diagnosing and resolving issues
└── scripts/
└── package.sh # Build linkly-ai.zip for upload
| File | Purpose |
|---|---|
SKILL.md |
Main instructions: environment detection, workflow, best practices |
references/cli-reference.md |
CLI commands, options, JSON output format |
references/mcp-tools-reference.md |
MCP tool parameters, response schemas, supported document types |
references/search-strategies.md |
Advanced query crafting, multi-round search, library scoping |
references/troubleshooting.md |
Connection issues, version mismatches, and diagnostic steps |
Compatibility
This skill follows the Agent Skills open standard and works with:
- Claude Code (Anthropic)
- OpenClaw
- Codex CLI (OpenAI)
- Any agent supporting the Agent Skills specification
Contributing
Contributions are welcome! Please open an issue or submit a pull request on GitHub.
License
Reviews (0)
Sign in to leave a review.
Leave a reviewNo results found