WIP simplifying load/startup

docs/architecture.md

@@ -85,7 +85,7 @@ Only canonical catalog resources are part of the runtime contract in this phase.
 
 ### Registry Loader
 
-The runtime composition includes a startup registry loader that reads packaged docs resources using `importlib.resources.files(...)` and `Traversable` APIs.
+Importing the package does not read or parse documentation. The MCP server and FastAPI application factories request the registry when constructing a runnable server, using packaged resources through `importlib.resources.files(...)` and `Traversable` APIs.
 
 Loader responsibilities:
 
@@ -94,12 +94,16 @@ Loader responsibilities:
 3. Build an in-memory registry keyed by `skill_id`.
 4. Fail fast for duplicate ids, missing markdown files, and broken reference mappings.
 
-Registry load failure is a startup error, not a partial runtime warning.
+The immutable registry is cached for the process lifetime. Each Uvicorn worker constructs and retains its own registry because worker processes do not share Python objects. Registry load failure is a server-factory startup error, not a package-import error or partial runtime warning.
 
 ### Content Sources
 
 Content is authored in markdown under `docs/` and managed as long-form reference material. Skill documents and companion references now live under `docs/skills/`, while project-authored pages remain alongside them in the docs tree. Resource handlers expose the same authored documents through stable resource URIs.
 
+The repository root `docs/` directory is the only authored source. The `src/personal_mcp/docs` path is a relative symlink to that directory for source-checkout and editable-install workflows; it is not a second content tree and packaging does not depend on traversing it.
+
+For wheel builds, [Hatchling forced inclusion](https://hatch.pypa.io/latest/config/build/#forced-inclusion) maps the root `docs/` tree to `personal_mcp/docs/`. The wheel therefore contains regular resource files at that destination rather than a symlink. Runtime registry loading uses [`importlib.resources.files`](https://docs.python.org/3/library/importlib.resources.html#importlib.resources.files) and `Traversable` operations from the `personal_mcp` package anchor, so it does not depend on the repository layout or current working directory.
+
 ### Static Docs Surface
 
 Static docs are built directly from two markdown source streams:
@@ -109,6 +113,8 @@ Static docs are built directly from two markdown source streams:
 
 The merged docs tree is built by Zensical into static files and served by the FastAPI app.
 
+Generated `site/` files are deployment assets for the human-facing static site. They are separate from the authored Markdown resources packaged under `personal_mcp/docs/`.
+
 ## Data Flow
 
 ```mermaid

docs/authoring.md

@@ -35,6 +35,14 @@ docs/
 			*.md
 ```
 
+## Source Tree Ownership
+
+Edit content only under the repository root `docs/` directory. The `src/personal_mcp/docs` path is a relative symlink provided so package-oriented tooling and editable installs see the same files; do not replace it with copied content or author files through a second tree.
+
+[Hatchling forced inclusion](https://hatch.pypa.io/latest/config/build/#forced-inclusion) projects root `docs/` into `personal_mcp/docs/` when building the wheel. Installed code reads that destination through [`importlib.resources`](https://docs.python.org/3/library/importlib.resources.html), while Zensical continues to build the human-facing site directly from root `docs/`.
+
+Package import does not load these resources. A runnable MCP or FastAPI server loads and validates them when its factory runs, then caches the immutable registry for that process. Restart initialized development or worker processes after changing authored Markdown.
+
 ## Authoring Principles
 
 1. Keep Markdown as the canonical source and avoid duplicating content into alternate metadata files.