How to use this reference
The Language Reference specifies the authoring surface of Aggrega — the YAML schemas, the shared keyword vocabulary, the Go template system, the provider contracts, the CLI commands, the compiler’s error catalog, and end-to-end worked examples. Read the chapters below in order for first-time orientation, or jump directly into a subdirectory when looking up a specific entity, keyword, provider, CLI command, or error.
To install the aggrega binary first, follow Install the binary; to have an AI assistant install and set it up for you, paste the prompt in Set up Aggrega with your AI assistant. Before you build scripts or a catalog on Aggrega, read Compatibility for what an upgrade may and may not change.
Before reading the Reference, ground yourself in the mental model the architecture chapters establish. The Four Domains and Team Topologies chapter explains how Aggrega separates Developer Intent, Platform Catalog, Platform Ops, and Platform Operator concerns into distinct authoring surfaces; the Entity Taxonomy chapter names the entities (Aggregate, Component, Feature, Recipe, FeatureRecipe, Plugin, Manifest, Release, SchemaLibrary, MappingLibrary, WorkloadLibrary, ArtifactLibrary) and their roles across those domains. Every Reference chapter after this one assumes the reader has that shape.
Use this Reference in two modes. For onboarding, read sequentially: start with Naming Conventions for the instance-name pattern and casing rules, then walk through entity and vocabulary chapters to build up the full authoring contract. For lookup, jump directly to the subdirectory that matches the question — entities/ for per-entity field semantics, vocabulary/ for cross-entity keywords, templates/ for the Go template surface available to Recipe authors, providers/ for per-provider dispatch contracts, plugins/ for the plugin sockets and plugin authoring, schema/ for the JSON Schema of each kind, cli/ for command-line usage, errors/ for diagnostics, and examples/ for worked patterns.