Skip to content

Architecture

MAESTRO is the system at the top of a four-part stack — you use it through the MAESTRO CARE TUI and the care CLI. Each part owns one stage of a chain’s life.

flowchart LR
    User([You]) --> Care["MAESTRO"]
    Care -->|task| MAGE
    MAGE -->|CARL chain| Care
    Care -->|run| Result([Result])
    Care <-->|save / load| Memory[(GigaEvo Memory)]
    Care -->|evolve| Platform[GigaEvo Platform]
    Platform -->|winner| Memory
PartPackageStageRole
MAGEmmar-mageGenerationTurns a natural-language task into a CARL chain.
CARLmmar-carlChain formatThe format every chain is written in — typed steps, dependencies (DAG), and per-step context. See the CARL docs.
GigaEvo Memorygigaevo-clientPersistenceStores entities (chain / agent / agent_skill / memory_card), the library, run history.
GigaEvo Platform—EvolutionGenetic search over chains; accept-and-promote the winner.
  1. Generation — describe a task; MAGE plans a CARL chain.
  2. Execution — MAESTRO runs the chain and returns a result (with a token/cost trail).
  3. Persistence — in Production mode, MAESTRO saves the chain to Memory under a stable chain_id and records each run.
  4. Evolution — optionally, Platform runs a GA over the chain and the best individual is accepted back into the stable channel.

Generate Agent A → save it → generate B and C → return to A from the Library → re-run from the same task and context files → optionally evolve A and accept the best individual back into the stable channel.

MAESTRO imports every upstream module lazily — inside the function that needs it. So a minimal install still boots the CLI and TUI; a missing piece (e.g. the optional maestro-care[carl] extra, or an unconfigured Memory URL) surfaces as a friendly hint rather than a crash. Without Memory configured, Production mode auto-falls back to Ad-Hoc.