Summarizes, extracts, and rewrites content from files.
89M visits/mo+8.7%#2 Productivity & Office
Kerlig
A macOS AI writing assistant that proofreads, rewrites, summarizes, and drafts emails and Slack replies with cross-app integration and PDF/image chat.
$49Starting price
34KMonthly visits
25Popularity
About Kerlig
Kerlig is a desktop AI writing assistant designed for macOS users who need AI-powered writing help inside any application, including email, Slack, and Jira, without switching to a browser. It provides system-wide writing actions for proofreading, spelling and grammar fixes, rewrites, tone adjustments, summarization, translation, and structured outputs like converting text into a TypeScript interface. Users can choose from over 350 AI models via their own API keys from providers such as OpenAI, Anthropic, Google, Groq, and OpenRouter, or use local models like Apple Foundation Models, Ollama, and LM Studio. The tool supports chatting with attachments including PDFs, ebooks, documents, and images, with features like chat history, prompt editing, and response regeneration. Kerlig also offers shortcuts and automation through global hotkeys, built-in and custom actions, presets, and team deployment options such as remote CLI deployment and headless mode. The tool operates on a one-time license model with no subscription, though token usage is billed through connected providers. It emphasizes privacy by sending messages directly to providers and storing data encrypted locally on the Mac.
Key features
- System-wide writing actions for proofreading, grammar fixes, and rewrites
- Support for 350+ AI models via user API keys or local models
- Chat with PDFs, ebooks, documents, and images (vision capabilities)
- Global hotkeys and built-in/custom actions for automation
- Team deployment options including remote CLI and headless mode
- One-time license with no subscription, plus token usage billing
- Cross-app integration for email, Slack, Jira, and other macOS apps
- Structured output generation (e.g., TypeScript interfaces)
- Local data encryption and privacy-focused message routing
- Presets and prompt editing for reusable workflows
Use cases
- Drafting and refining professional emails and Slack replies
- Summarizing and analyzing documents or PDFs
- Automating repetitive writing tasks across applications
Pros
- Integrates AI writing assistance directly into macOS workflows without requiring a browser switch
- Supports over 350 AI models via user-provided API keys or local models like Apple Foundation Models, Ollama, and LM Studio
- Offers system-wide writing actions including proofreading, rewrites, tone adjustments, summarization, translation, and structured outputs
- Provides document analysis and chat capabilities with attachments such as PDFs, ebooks, and images
- Operates on a one-time license model with no subscription, though token usage is billed through connected providers
Cons
- Requires users to manage their own API keys for most models, adding setup complexity
- Limited to macOS users, excluding other operating systems
- Token costs are incurred through third-party providers, which may vary and add unpredictability to expenses
Frequently asked questions about Kerlig
What is Kerlig and how does it work?
Kerlig is a desktop AI writing assistant for macOS that integrates AI-powered writing actions directly into any application, such as email, Slack, or Jira. It operates system-wide, allowing users to proofread, rewrite, adjust tone, summarize, translate, or generate structured outputs like TypeScript interfaces without switching to a browser.
Who is Kerlig designed for?
Kerlig is designed for macOS users who frequently write emails, Slack messages, Jira tickets, or other professional communications and want to streamline their workflow with AI assistance. It suits individuals, professionals, and teams seeking a native, efficient AI tool integrated into their daily tasks.
What AI models does Kerlig support?
Kerlig supports over 350 AI models via user-provided API keys from providers like OpenAI, Anthropic, Google, Groq, and OpenRouter, as well as local models such as Apple Foundation Models, Ollama, and LM Studio. Users can switch between models with one click.
How does Kerlig handle privacy and data storage?
Kerlig sends messages directly to the chosen AI provider and stores data encrypted locally on the user's Mac. This ensures that sensitive information remains secure and private, with no data stored in the cloud by Kerlig itself.
What are the licensing options for Kerlig?
Kerlig operates on a one-time license model with no subscription fees. Users pay once for the license and only incur costs for tokens used through their connected AI providers. Licenses are available for individual users, professionals with multiple Macs, and teams.
Can Kerlig be used for team deployment?
Yes, Kerlig offers team deployment options, including remote CLI deployment and headless mode, allowing teams to integrate AI workflows across multiple Macs. Team licenses support up to 10 seats with priority support and additional features like remote management.
Kerlig traffic & engagement
kerlig.com · third-party estimates from Similarweb
Data as of Sep 2026Full AI website ranking →
Monthly visits · Sep 202634K-2.8% MoM12-month high 35K (Aug 2026)
Rank among AI websites#3,091▼ 28 vs last monthof 8,652 tools on aitoolnavio™
Top marketUnited States14.8%Top 5 countries bring 45% of visits
Average visit14sTime on site per visit
Pages per visit1.68Average pages viewed
Bounce rate40%Leave after one page · lower is better
Similarweb global rank#937,701#1,183,472 in United States
Traffic trend
Last 4 months · grows by one month every month
- Jun 202626K
- Jul 202629K+11.1%
- Aug 202635K+18.6%
- Sep 202634K−2.8%
Top countries
Share of visits · estimated visits
Other countries: 54.7%
Visits sent by AI assistants
How people reach this site from other AI tools
9 a month
Compared with similar tools
Productivity & Office · Sep 2026
Traffic figures are third-party estimates, good for scale and trends; they can differ from a company's own numbers. Updated monthly.
