[P5] Add typed transformer lifecycle hooks

[KayleeWilliams]

Problem

Leadtype has a strong internal pipeline, but users who need custom behavior currently have to work around it or fork pieces. We need extension points that preserve the pipeline's stability.

Proposal

Add typed transformer lifecycle hooks around parsing, markdown flattening, search indexing, and LLM artifact generation.

Example:

export default defineDocsConfig({
  transformers: [
    {
      name: "api-area-metadata",
      afterFrontmatter(page) {
        return {
          ...page,
          frontmatter: {
            ...page.frontmatter,
            apiArea: inferApiArea(page.filePath),
          },
        };
      },
      beforeSearchChunk(chunk) {
        return boostApiSymbols(chunk);
      },
    },
  ],
});

Potential hooks:

• beforeParse
• afterFrontmatter
• afterMdxAst
• afterFlattenMarkdown
• beforeSearchIndex
• beforeLlmsTxt
• beforeAgentsMd

Core Behavior

• Hooks operate on Leadtype pipeline objects, not framework request/response objects.
• Hooks run in deterministic order.
• Hook errors include transformer name and source path.
• Existing remark plugin behavior remains compatible.

Framework Integration Notes

Docs should show framework-specific consumption patterns:

• SvelteKit: derive route/layout metadata for load.
• Nuxt: enrich pages for navigation/sidebar fields.
• Astro: align generated metadata with Content Collections conventions.
• Next/Fumadocs: derive sidebar/search metadata.

Acceptance Criteria

• Hooks are typed and documented.
• Hooks can customize frontmatter-derived metadata.
• Hooks can customize search chunks without replacing the search generator.
• Hooks can customize LLM artifacts without replacing the full generator.
• Hook failures include transformer name and file path.
• Pipeline output remains deterministic.

[pullfrog]

  ^{｜ Build this ➔ ｜ Make a plan ➔}

[coderabbitai]

🔗 Related PRs

#6 - Add docs authoring components [merged]
#23 - Polish FileTree, Prompt, and TopicSwitcher in docs example [merged]
#32 - Shrink leadtype dependency closure [merged]
#35 - Add v1 headless integration surface and polish docs [merged]
#37 - Fix type table source root resolution [merged]

---

📝 Issue Planner

_{Check the box below or use the @coderabbitai plan command to generate an implementation plan and prompts that you can use with your favorite coding assistant.}

• Create Plan

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!