Symfony Spot Hit Notifier Laravel Package

Product Decisions This Supports

• Build vs. buy for mail delivery plumbing. This package is a thin Laravel mail transport adapter (it plugs a custom/third-party mail provider into Laravel's Mail facade). The decision it informs is whether to adopt it, wrap it, or implement the equivalent driver in-house — the surface area is small enough that "build" is a realistic option, so the package must earn its place by saving ongoing maintenance, not just initial effort.
• Feature feasibility for transactional messaging. Any roadmap item depending on a specific email provider (transactional sends, templates, delivery webhooks) can be scoped against this package's transport layer instead of a bespoke integration.
• Configuration contract ownership. Release 0.0.2 ("the Transport host should not contain https") signals that the package's config semantics are still being settled: host is a bare hostname, not a URL. That is effectively a breaking change for any existing config that passed https://smtp.example.com, and it tells us the validation/normalization of that value belongs on our side, not the vendor's.
• Release-risk tiering. Version 0.0.x means no API stability guarantee. This is a "pilot / non-critical path" dependency, not something to place behind revenue-critical mail without an adapter layer and pinned versions.

When to Consider This Package

Adopt (with guardrails) when:

• You need a Laravel-native mail transport for a provider that Laravel's first-party mailers don't cover, and you want to keep using Mail::to(...) rather than hand-rolling an HTTP client per feature.
• You can pin the exact version and treat upgrades as reviewed changes — 0.0.2 is a fix that will silently break configs containing a scheme, which is normal for pre-1.0 but must be planned for.
• You will wrap it in your own config class or config validator that strips http(s):// and trailing slashes from the host, so the footgun can't recur on the next minor release.
• The alternative is writing and maintaining the driver yourself, and the provider's API is stable enough that the adapter's churn is low.

Look elsewhere when:

• You only need SMTP or a provider with an official Laravel/Symfony transport — use the built-in driver; this package adds a dependency for no gain.
• You require vendor support SLAs, a security-response commitment, or a stable public API — none of which a 0.0.x package can offer.
• Mail is on the critical path (password resets, receipts, compliance notices) and you can't absorb an unannounced config-semantics change.
• You need broad provider coverage or features beyond a single transport (webhooks, bounce handling, templating) — this appears to be a narrow adapter, and those concerns stay yours.

How to Pitch It (Stakeholders)

For executives: This is a small, low-cost integration library that lets the product send email through a specific provider using Laravel's standard mail interface. It's currently at version 0.0.2 — pre-stable — so I'd recommend it for a pilot or a non-critical notification path, not for anything customer-critical yet. The 0.0.2 release fixed a configuration gotcha around how the mail server address is specified; it's a good sign the maintainer is iterating quickly, but also a reminder that the interface isn't frozen. Cost of adoption and cost of backing out are both low, which is why it's worth trying now rather than building our own.

For engineering: It's a thin transport adapter — the kind of thing one of us could write in a day, so the value is maintenance amortization, not capability. Two concrete asks if we adopt: (1) pin the version and review upgrades, since 0.0.2 changed config semantics (host must be a bare hostname — a scheme like https:// must be stripped) in a way semver permits pre-1.0; (2) put our own normalization/validation in front of the config so a future release can't reintroduce that class of bug. Keep it behind our own small interface so swapping to a first-party driver or an in-house implementation stays a contained change, and revisit the decision at 1.0 or if the maintainer goes quiet.