Add TODO.md with planned features and refactoring
Lint Checks / Run linter (push) Failing after 0s
Details
Lint Checks / Run linter (push) Failing after 0s
Details
- Update stack command for continuous deployment workflow - Separate deployer from CLI - Separate stacks from orchestrator repo Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>helm-charts-with-caddy
parent
8d9682eb47
commit
811bbd9db4
|
|
@ -0,0 +1,16 @@
|
||||||
|
# TODO
|
||||||
|
|
||||||
|
## Features Needed
|
||||||
|
|
||||||
|
### Update Stack Command
|
||||||
|
We need an "update stack" command in stack orchestrator and cleaner documentation regarding how to do continuous deployment with and without payments.
|
||||||
|
|
||||||
|
**Context**: Currently, `deploy init` generates a spec file and `deploy create` creates a deployment directory. The `deployment update` command (added by Thomas Lackey) only syncs env vars and restarts - it doesn't regenerate configurations. There's a gap in the workflow for updating stack configurations after initial deployment.
|
||||||
|
|
||||||
|
## Architecture Refactoring
|
||||||
|
|
||||||
|
### Separate Deployer from Stack Orchestrator CLI
|
||||||
|
The deployer logic should be decoupled from the CLI tool to allow independent development and reuse.
|
||||||
|
|
||||||
|
### Separate Stacks from Stack Orchestrator Repo
|
||||||
|
Stacks should live in their own repositories, not bundled with the orchestrator tool. This allows stacks to evolve independently and be maintained by different teams.
|
||||||
Loading…
Reference in New Issue