Discover MCPs & agents
Loading MCPs and agents…
Loading MCPs and agents…
Lybrary can be dropped into any AI-assisted development workspace. It gives humans and agents a shared, well-structured home for Markdown documents, prompts, specs, and playbooks so context stays organized across experiments, sessions, and handoffs.
From the repo.
Lybrary Experimental packages the evolving Lybrary knowledge system for AI-assisted teams. This top-level repository keeps the experimental lybrary-experimental/ workspace available while the platform remains under active research and documentation.
Note: This is the experimental version of Lybrary. Lybrary Team α is actively researching and documenting the system while making it available as open-source "Early access."
lybrary-experimental/README.md for a deep orientation to the current folder layout, roles, and normalization workflow.lybrary-experimental/lybrary/ into your project or agent workspace to seed the shared context space.lybrary-experimental/lybrary/idea_space/ so your project team can iterate.lybrary-experimental/lybrary/lybrary.md as well as lybrary_rules.mdlybrary-experimental/README.md — source-of-truth documentation for the experimental Lybrary, updated frequently as we learn.lybrary-experimental/lybrary/ — portable directory of rules, knowledge, modules, and agent intake folders ready to drop into your workflows.lybrary-experimental/changelog-0.md — running log of early changes and milestones.Lybrary Team α is continuing to refine usage guidance, improve the directory scaffolding, and expand examples while analyzing results from real-world sessions. In parallel, we are drafting a formal paper about the system to share our methodology and findings.
Expect regular updates to the README, changelog, and module templates as we validate new practices. Early adopters are encouraged to experiment, document what works, and surface gaps so the research can incorporate your experience.
Open issues and share notes in this repository, your insights help steer the research agenda and shape the eventual stable release!
Yes, we know there are a lot of README.md files, they are clearly titled on line 1 for AGENTS, and we are currently considering switching to a different naming convention for the per-directory README files.
Apologies in advance for any trouble with human-readability on that.
Replace {MCP_ENDPOINT_URL} with this MCP’s endpoint URL (from its repo or docs above). No API key — you connect directly.
Tool
OS
Config file: ~/.cursor/mcp.json
{
"mcpServers": {
"mcp-server": {
"url": "{MCP_ENDPOINT_URL}"
}
}
}Paste into mcpServers in the config file. Restart Cursor after saving.
If this MCP is also published on mcpchannel.ai, you can subscribe from Browse and use the gateway config there instead.