Sklm
<p> Centralize, scope, and sync skills for every AI agent </p> <p> <a href="https://www.producthunt.com/products/sklm?utm_campaign=producthunt-atom-posts-feed&utm_medium=rss-feed&utm_source=producthunt-atom-posts-feed">Discussion</a> | <a href="https://www.producthunt.com/r/p/1172208?app_id=339">Link</a> </p>
Last verified:
What is Sklm?
Sklm is an AI Agent Skills Manager that helps you centralize, scope, and sync skills for AI agents from the command line. It is designed to solve the common problem of wanting some skills to be available everywhere while keeping others limited to a specific project.
The tool installs skills once into a global store at ~/.sklm/ and then lets you activate only the ones you want in a given workspace. That means each project can expose a tailored set of skills without cluttering the agent’s configuration or duplicating files across repositories.
Sklm also supports registry discovery, so you can index local folders or git repositories as searchable catalogs of skills. The website says it automatically syncs agent config directories when you add or remove skills, reducing manual copying and keeping the setup consistent.
It is built for people who work with multiple AI coding agents and want one workflow across them. The site highlights support for 30 agents, including OpenCode, Claude Code, Cursor, Windsurf, Gemini CLI, Cline, Amazon Q, Codex CLI, Continue, and GitHub Copilot explicit support.
Sklm pricing
Pricing model: Freemium
The website does not list any pricing, paid plans, or subscription tiers. Based on the site content, Sklm is presented as a free CLI tool with a global store, registry support, project scoping, and migration features, but no monetization details are shown on the homepage.
Sklm pros
- Centralizes skills in one global store
- Supports per-project skill scoping
- Uses symlinks for clean activation
- Automatically syncs agent config directories
- Reduces manual file copying
- Supports 30 AI agents
- Works with OpenCode
- Works with Claude Code
- Works with Cursor
- Works with Windsurf
- Works with Gemini CLI
- Works with Cline
- Works with Amazon Q
- Works with Codex CLI
- Works with GitHub Copilot explicit mode
- Indexes local directories as registries
- Indexes git repositories as registries
- Searches registries by keyword
- Has a global cache/store architecture
- Supports migration of existing local skills
Sklm cons
- CLI-first workflow only
- Requires project-level activation step
- Depends on symlink-based setup
- Needs filesystem access to agent config folders
- Best suited to supported agents only
- Explicit support required for some agents
- Requires learning its init/install/add/migrate flow
- Global store can add another management layer
- Registry setup may be extra work for simple use cases
Frequently asked questions about Sklm
What does Sklm do?
Sklm is a skills manager for AI agents that lets you install skills once, keep them in a global store, and activate only the ones needed for each project. It is meant to make skills easier to organize across multiple agents without manually copying files around.
How does Sklm keep project skills isolated?
It uses per-project scoping with symlinks, so a workspace only sees the skills you explicitly activate there. The homepage emphasizes that this avoids polluting an agent’s configuration while still keeping skills available globally.
Where are skills stored?
The site says skills are installed into ~/.sklm/ as a global store. The workspace then links into that store through a local .sklm/ directory.
What commands are part of the workflow?
The website shows a four-step flow: init, install, add, and migrate. Init prepares Sklm in a project, install adds skills to the global store, add exposes them in the current project, and migrate brings existing local skills under Sklm management.
Which AI agents does Sklm support?
The homepage says Sklm recognizes 30 AI agents. Examples shown include OpenCode, Claude Code, Cursor, Windsurf, Gemini CLI, Cline, Amazon Q, Codex CLI, Continue, Qwen Code, and GitHub Copilot explicit support.
Can Sklm discover skills from repositories?
Yes. The site says it can index local directories or git repositories as registries, then search them by keyword across all your registries.
Does Sklm sync changes automatically?
Yes. The homepage states that sklm add and sklm rm automatically sync the agent’s config directory, so you do not have to copy files by hand after every change.
Can Sklm migrate skills I already have?
Yes. Migration is listed as one of the main pipeline steps, and the website explicitly says you can migrate existing local skills into Sklm’s management.
Is Sklm tied to one agent vendor?
No. The site positions it as a multi-agent tool that works across many different agent ecosystems and config locations, rather than being locked to one provider.
What problem is Sklm trying to solve?
It addresses the tension between making skills globally available and keeping them scoped to individual projects. The homepage says Sklm solves this by installing once globally and activating per project only what you want.