Skip to main content

One repo-local source of truth for every agent you use.

terminal

# Install agent-layer

$ brew install conn-castle/tap/agent-layer

# Initialize agent-layer in your repo

$ al init

# Launch your agent (sync happens automatically)

$ al claude

Works with your favorite agents

Claude CodeGemini CLIOpenAI Codex CLIGitHub CopilotAntigravity

Why Agent Layer?

Eliminate duplicate configuration work and deliver faster with consistent agent setup.

Single Source of Truth

Edit .agent-layer/ once. Skills align with the agentskills.io specification, and every client gets its native config.

Sync on Run

Configs regenerate every launch. No manual steps, no drift.

Consistent Approvals

Define safety rules once. Every agent respects the same boundaries.

MCP Everywhere

Configure MCP servers in one place. They appear in every client.

Version Pinning

Lock your team to specific versions. Everyone stays aligned.

Local & Fast

No telemetry. Network is used only for update checks, version downloads (pinning), and MCP servers you enable. See Security for details.

Ready to simplify your agent setup?

Install Agent Layer in under a minute. Works on macOS and Linux.

Support the project

If Agent Layer is valuable for your team, please consider starring the repository. Stars improve discovery for new users evaluating the project.