Overview

Topic-focused documentation for @weave-kit/engine. New to the engine? Read Getting started first, then MCP to put an agent on your data. The project pitch and quick start live in the repository README.

Guides

• Getting started — scaffold, migrate, run, and consume the API
• Schema guide — fields, relations, computed fields and validation
• RBAC — roles, permissions and row scopes
• Formulas — operators, functions, aggregations and null semantics
• Audit — the immutable event log and diff replay
• CLI reference — every weave command
• MCP — expose objects as agent tools, per identity
• Script subsystem — sandboxed *.server.js lifecycle hooks
• Custom tools, guardrails & audit replay — the open contract
• Quotas — usage budgets consumed by tools and scripts
• Inbound events — signed webhooks mapped to engine actions

Operations

• Deploying with Docker — containerize the engine + PostgreSQL
• Reverse proxy + TLS — Caddy / Nginx and streaming caveats

Practices

• Integrating an existing CRM with MCP — start from live tables
• Plugging in the customer's own user store — resolver-driven identity
• Connecting MCP hosts — Claude Desktop, Cursor, gateways
• Connect an agent — local first run in five minutes

Reference

• Public API & dependency budget — entry-point tiers and extension points