JSPM

@blacklake-systems/surface-cli

0.2.15
  • ESM via JSPM
  • ES Module Entrypoint
  • Export Map
  • Keywords
  • License
  • Repository URL
  • TypeScript Types
  • README
  • Created
  • Published
  • Downloads 6
  • Score
    100M100P100Q8395F
  • License MIT

BlackLake Surface — control plane for your AI agents (cloud or self-hosted)

Package Exports

    This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (@blacklake-systems/surface-cli) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

    Readme

    @blacklake-systems/surface-cli

    ⚠️ Deprecated. This package is now part of the unified blacklake npm package. Install blacklake and use npx blacklake serve for the same local API + dashboard + SQLite. See the migration doc for sed-style search-and-replace examples. This package will continue to ship as a thin re-export through the next two minor versions.

    BlackLake — AI control infrastructure and analytics. This package historically shipped the local CLI; it now re-exports through the unified blacklake package.

    Surface is the control layer for AI actions. Use the CLI when Codex, Claude, Cursor, CI jobs, backend agents, or shell commands are about to touch real systems and you want the same policy, approval, cost, and signed-receipt model locally.

    Install

    npx @blacklake-systems/surface-cli

    This starts the local Surface API and console, backed by ~/.blacklake/blacklake.db. Local is the fastest private first run; the hosted console is for shared policies, approvals, budgets, exports, and team visibility.

    Commands

    blacklake
    blacklake --help
    blacklake --version
    blacklake mcp
    blacklake blx <cmd...>
    blacklake demo depth-with-approval

    MCP client integration

    For cloud users, point MCP-compatible tools at the hosted gateway:

    {
      "mcpServers": {
        "blacklake": {
          "command": "npx",
          "args": [
            "-y",
            "mcp-remote@latest",
            "https://api.blacklake.systems/mcp",
            "--header",
            "Authorization: Bearer YOUR_BL_USER_KEY"
          ]
        }
      }
    }

    Register real MCP services in the console first, for example:

    • cloudflare-docs
    • engineering-github-repositories
    • product-linear-workspace
    • company-notion-workspace
    • jira-confluence-workspace

    For local users, run blacklake mcp as a stdio bridge from your MCP client and put upstream MCP services in ~/.blacklake/mcp-config.json.

    Shell command governance

    Use blx when an AI-assisted shell workflow is about to run a consequential command:

    blacklake blx git push origin main
    blacklake blx gh pr merge 123 --squash
    blacklake blx gcloud run deploy api
    blacklake blx terraform apply
    blacklake blx kubectl apply -f deploy.yaml

    Read-only/build commands are allowed through. Write/deploy/destructive commands are classified and governed before execution.

    Demo

    blacklake demo depth-with-approval

    Creates a small Depth workflow that pauses on a Surface approval.

    Product boundaries

    BlackLake controls the paths you route through it: MCP gateway, SDK, GitHub Action, blx, cloud audit ingest, and Depth workflows. It does not invisibly intercept every process on the machine or replace cloud IAM.