Blekline is Layer 4 governance in the AI Enablement Stack — mask, enforce, and audit agent traffic before it reaches models and L1 sandboxes.
1. Workspace token
Create an API key in Admin → API keys (or Operations → MCP) with scopes mask:write and events:write. Use this token in MCP config and API calls.
2. Install MCP server (npm — recommended)
npm install -g @blekline/mcp-server@0.3.0
Set your workspace token:
export BLEKLINE_WORKSPACE_TOKEN="blw_..."
export BLEKLINE_API_URL="https://app.blekline.com"
3. Connect your client
| Priority | Client | Config |
|---|---|---|
| P0 | Cursor | .cursor/mcp.json — see Cursor |
| P1 | Claude Desktop | Claude Desktop |
| P2 | Codex | Codex |
| P3+ | Continue, Copilot, OpenHands, Cody | Agent clients hub |
Generate configs from Operations → MCP in the app, or use the open-source repo scripts if you clone blekline-oss.
4. Verify
In Cursor: Settings → MCP → blekline should show green with MCP tools (blekline_mask_prompt, etc.).
Optional API smoke:
curl -sS https://app.blekline.com/api/ready
Running agents in sandboxes?
If your agents execute inside an L1 sandbox (Daytona, Modal, E2B, Cloudflare, Vercel Sandbox), run @blekline/mcp-proxy as a sidecar alongside the provider MCP server. The proxy intercepts tool calls before they reach the sandbox.
See Sandbox providers for L1 wiring. For model APIs (Azure, OpenRouter, Bedrock), see Model providers.
Contributors (monorepo)
If you develop Blekline from source:
git clone https://github.com/Blekline/blekline-oss.git
cd blekline-oss
pnpm install
pnpm build:packages
cp .env.example .env
pnpm generate:mcp-configs
pnpm demo:mcp-smoke
Optional: browser extension
The browser extension masks prompts in web UIs. MCP is the primary path for Cursor, Claude Desktop, and Codex — the extension is not required.
Next steps: AI Enablement Stack · Architecture · Agent clients · Model providers · Open workspace