dryrun

PostgreSQL schema intelligence from a committed snapshot. No connection, no credentials. Capture the schema once, and your AI assistant, IDE, and CI get full schema awareness — tables, indexes, constraints, statistics — without a connection string on any developer machine.

This npm package is a thin launcher. The real binary is written in Go and ships as a platform-specific dependency that npm selects automatically.

Use as an MCP server

No install needed. Add to your MCP client config:

{
  "mcpServers": {
    "dryrun": {
      "command": "npx",
      "args": ["-y", "@boringsql/dryrun", "mcp-serve"]
    }
  }
}

npx fetches the package on first run and caches it; subsequent launches are local. Pin a version for reproducibility: "@boringsql/[email protected]".

Or let dryrun write the config. setup detects Claude Code, Cursor, Codex, and Zed, and adds a directive to AGENTS.md/CLAUDE.md so the agent checks the schema before writing SQL:

npx @boringsql/dryrun setup

The server reads the newest snapshot from .dryrun/history.db in the current project. Without one, its tools answer that no schema is loaded: capture with dryrun init --db "$DATABASE_URL", or pull a snapshot a teammate pushed with dryrun snapshot pull. It is picked up on the next tool call, no restart needed.

CLI

npx @boringsql/dryrun --help

Or install globally: npm i -g @boringsql/dryrun then dryrun --help.

Supported platforms

darwin-arm64, darwin-x64, linux-x64, linux-arm64, win32-x64. On other platforms, grab a prebuilt binary from the GitHub Releases or install via Homebrew (brew install boringsql/boringsql/dryrun).

More

• GitHub README — quickstart, team snapshot sharing, OCI registry push/pull
• Tutorial — offline, online, and multi-node workflows with the full tool reference
• boringsql.com/products/dryrun — project page