Maintainer Laravel Package

Product Decisions This Supports

• Release automation and changelog quality: The package now enforces strict structured output for AI-generated changelog entries, constraining each entry’s type to the Conventional Commit set (feat, fix, docs, style, refactor, perf, test, build, ci, chore, revert). This directly supports decisions about automating release notes without accepting malformed or non-standard categories.
• Build vs. buy for release tooling: If your team already uses Laravel and AI-assisted changelog generation, this package reduces the need to build custom validation around provider output. The new fail-fast behavior and richer diagnostics make it more viable as a maintained dependency rather than a homegrown script.
• Roadmap prioritization for developer experience: The release improves error messages with commit hash, invalid type, and allowed values. This supports investing in automation that surfaces actionable failures early, rather than requiring manual triage after a release pipeline breaks.
• Adoption of Conventional Commits: The strict enum constraint reinforces a Conventional Commit taxonomy across AI-generated entries. PMs can use this to align release notes, semantic versioning signals, and downstream tooling around a consistent changelog schema.
• Risk management for AI-dependent workflows: The change may break workflows that previously accepted non-compliant AI output. This supports decisions to audit AI provider behavior, update prompts or provider configuration, and treat changelog generation as a validated pipeline step rather than best-effort output.

When to Consider This Package

Consider this package when:

• Your team uses Laravel/PHP and wants AI-assisted release changelog generation integrated into an existing release workflow.
• You already follow, or want to enforce, Conventional Commits and need generated changelog entries to use a fixed, validated type set.
• You value fail-fast validation over silently accepting malformed AI output, especially where changelog entries feed release notes, versioning decisions, or customer-facing communications.
• You want clearer diagnostics when an AI provider returns invalid data, including the offending commit, invalid value, and allowed values.
• You are comfortable with AI provider dependency and want the package to request strict structured output from supported providers.

Look elsewhere when:

• Your workflow relies on non-Conventional Commit categories or custom changelog taxonomies that cannot be mapped to the allowed type enum.
• Your AI provider or internal workflow cannot reliably return strict structured output, and a fail-fast release abort would be unacceptable.
• You need fully deterministic, non-AI changelog generation or a highly customized release-notes format outside the package’s supported schema.
• You are not using Laravel/PHP or do not want to introduce AI provider dependencies into the release process.
• You require backward compatibility with previously permissive AI outputs and cannot update prompts, provider settings, or validation handling.

How to Pitch It (Stakeholders)

For executives: This release makes AI-generated changelogs more reliable and governable. Instead of accepting whatever an AI provider returns, the package now enforces a strict, Conventional Commit-compatible schema and fails early with clear diagnostics when output is invalid. That means less manual cleanup, more consistent release communications, and fewer silent quality issues in the release pipeline.

For engineering: The package now requests strict structured output from AI providers, runs the changelog agent in strict mode, and validates entries[].type against the allowed enum. Invalid values cause generate() to throw a RuntimeException containing the commit hash, the invalid value, and the expected types. Tests cover strict schema requests and rejection of unknown, null, or non-string types.

For product and release management: This is primarily a stricter validation change. Teams using AI-generated changelogs should confirm their provider and prompts return only supported type values. Workflows that relied on looser output will now fail early, which is intentional: it prevents non-compliant entries from reaching release notes and gives teams a precise diagnostic to fix the source.