Skip to content

Anvil

Anvil

Anvil is a declarative, provider-aware execution engine for running Python tasks across cloud and GitHub target fleets.

Describe the work in YAML, keep task logic in plain Python modules, and let the engine handle authentication, target discovery, dependency ordering, bounded concurrency, and structured results. AWS, Azure, GCP, and GitHub providers ship with Anvil, and extension packages can add providers, tasks, and processors.

Why Anvil?

Anvil is built for teams that need repeatable workflows to run consistently across provider-specific execution targets and locations. Common use cases include inventory, validation, enforcement, cleanup, security auditing, and reporting.

Core Capabilities

  • Declarative orchestration with reusable YAML configuration.
  • Provider-aware targets for AWS accounts, Azure subscriptions, GCP projects, and GitHub organizations or repositories.
  • Explicit bounded concurrency for configured targets, resolved execution targets, and regions or locations.
  • Shared provider discovery, authentication, and session reuse during a run.
  • Task isolation through plain Python task modules with a run(...) function.
  • Universal and provider-specific tasks plus extension-package components.
  • Structured task, execution-target, configured-target, and engine results.
  • Dry-run behavior, dependency ordering, fail-fast controls, and rerun support.

Installation

The base package includes AWS support:

pip install anvil

Install the SDK extras needed by the configured providers:

pip install "anvil[azure]"
pip install "anvil[gcp]"
pip install "anvil[github]"

Repository Template

Create a dedicated task repository with the foundry-anvil-template. The template provides a ready project layout for custom tasks, YAML examples, validation, and CI outside of the main Anvil repository.

Next

  • Read the execution model for how Anvil prepares targets, authenticates, resolves provider targets, and schedules task work.
  • Read the CLI reference for the main command groups and result commands.
  • Read the task contract for custom task discovery, validation, and the run(...) interface.
  • Read configuration for schema v2 provider targets, selection, location, concurrency, and metadata concepts.
  • Read examples for the repository template, GitHub Actions, and standalone multi-account script template.