René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"[Feature]: Promote `plugin` to Stable — freeze the hook-bus / DI / directive APIs","articleBody":"**Pre-flight checklist**\n- [x] I searched existing issues and did not find a duplicate request.\n- [x] I explained the problem or use case, not just the desired API.\n\n**Problem or use case**\n`plugin` is **Beta** and is the extensibility contract the whole ecosystem would build on — yet its surface is one minor old: the hook bus, DI container, async install, namespaced directives, and `unuse`/`uninstall` all arrived in 1.14.0. Third-party plugin authors cannot target a moving extension API; the contract must be frozen before an ecosystem can form.\n\n**Proposed solution**\nFreeze the plugin API (hook bus, DI container, install/uninstall lifecycle, namespaced directive registration) for at least one minor cycle and publish a plugin-author guide documenting the lifecycle, hook timing, DI resolution rules, and directive-namespacing conventions. Add coverage for install/uninstall symmetry (no leaked hooks/directives after `uninstall`). Exit criteria: frozen surface, author guide, tested lifecycle symmetry.\n\n**Possible API or UX shape**\n```ts\nimport { definePlugin } from \"@bquery/bquery/plugin\";\n\nexport default definePlugin({\n  name: \"my-plugin\",\n  async install(ctx) {\n    ctx.provide(\"token\", service);\n    ctx.directive(\"my:thing\", handler);\n    ctx.hook(\"view:updated\", fn);\n  },\n  uninstall(ctx) { /* must cleanly remove all of the above */ }\n});\n```\n\n**Alternatives considered**\nPromoting before freezing — would strand early plugin authors on a changing contract. Expanding the hook surface further first — fine while Beta, but it must stop for a cycle to earn Stable.\n\n**Relevant area**\n`plugin`\n\n**Additional context**\nA stable `plugin` API is a prerequisite for any community ecosystem, which is itself one of bQuery's adoption gaps versus the established frameworks. Lifecycle symmetry (clean `uninstall`) is the highest-risk area to test.\n","author":{"url":"https://github.com/JosunLP","@type":"Person","name":"JosunLP"},"datePublished":"2026-06-26T21:25:47.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":0},"url":"https://github.com/145/bQuery/issues/145"}