mainsequence 8.1.20.dev8__tar.gz → 8.1.20.dev9__tar.gz

{mainsequence-8.1.20.dev8/mainsequence.egg-info → mainsequence-8.1.20.dev9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mainsequence
-Version: 8.1.20.dev8
+Version: 8.1.20.dev9
 Summary: Main Sequence SDK 
 Author-email: Main Sequence GmbH <dev@main-sequence.io>
 License-Expression: LicenseRef-MainSequence-Proprietary

{mainsequence-8.1.20.dev8 → mainsequence-8.1.20.dev9}/agent_scaffold/skills/data_access/exploration/SKILL.md

@@ -60,9 +60,9 @@ This skill discovers and identifies data. It does not publish, redesign, or defi
 ## Read First
 
 1. `AGENTS.md`
-2. `docs/cli/index.md`
-3. `docs/knowledge/time_index_table_updates.md`
-4. `docs/knowledge/meta_tables/index.md` when the exploration touches row-oriented storage
+2. <https://mainsequence-sdk.github.io/mainsequence-sdk/cli/>
+3. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/time_index_table_updates/>
+4. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/meta_tables/> when the exploration touches row-oriented storage
 
 ## Inputs This Skill Needs
 

{mainsequence-8.1.20.dev8 → mainsequence-8.1.20.dev9}/agent_scaffold/skills/data_publishing/meta_table_migrations/SKILL.md

@@ -17,16 +17,13 @@ platform-managed application tables outside the migration workflow.
 
 ## Read First
 
-1. `docs/knowledge/meta_tables/migrations.md`
-2. `docs/knowledge/meta_tables/api.md`
-3. `docs/adr/0020-metatable-migration-artifact-registry.md`
-4. `docs/adr/0026-sdk-owned-migration-scaffolding.md`
+1. `AGENTS.md`
+2. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/meta_tables/migrations/>
+3. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/meta_tables/api/>
 
-`docs/knowledge/meta_tables/migrations.md` is the authoritative SDK reference for
-provider behavior and migration contracts. The runnable beginner migration
-walkthrough is maintained in the separate canonical tutorial CodeRepository at
-<https://github.com/mainsequence-projects/mainsequence-sdk-tutorial>; read it
-only when a runnable end-to-end example is needed.
+The pages above are the authoritative reference for provider behavior and
+migration contracts. This skill states the contract it needs; read those pages
+for the full provider and catalog-binding detail.
 
 ## Required Decisions
 
@@ -103,6 +100,12 @@ platform-managed, Alembic-managed reservations and are validated before HTTP.
 
 - Keep Alembic as the schema migration engine; do not build custom operation
   lists or fake migration payload formats.
+- These are intentionally unsupported and must not be reintroduced: SDK-managed
+  migration artifact table models, packaged migration rows,
+  `load_packaged_migration`, `sync_packaged_migration`,
+  `build_migration_registry_row`, custom `operations()` migration modules, and
+  operation names such as `add_column` or `create_index`. Use provider-selected
+  Alembic revisions and Alembic-rendered SQL only.
 - Keep provider scope explicit. Do not scan all imported models or installed
   packages.
 - For one-model or configured dynamic providers, use `metadata_for_models(...)`

{mainsequence-8.1.20.dev8 → mainsequence-8.1.20.dev9}/agent_scaffold/skills/data_publishing/meta_tables/SKILL.md

@@ -49,19 +49,16 @@ If the user is still in the discovery process and does not yet know what data ex
 
 ## Read First
 
-1. `docs/knowledge/meta_tables/index.md`
-2. `docs/knowledge/meta_tables/sqlalchemy.md`
-3. `docs/knowledge/meta_tables/compiled_sql.md`
-4. `docs/knowledge/meta_tables/migrations.md`
-5. `docs/knowledge/meta_tables/api.md`
-6. For migration work, use
+1. `AGENTS.md`
+2. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/meta_tables/>
+3. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/meta_tables/sqlalchemy/>
+4. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/meta_tables/compiled_sql/>
+5. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/meta_tables/migrations/>
+6. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/meta_tables/api/>
+7. For migration work, use
    `.agents/skills/mainsequence/data_publishing/meta_table_migrations/SKILL.md`
 
-The knowledge pages above are the authoritative SDK reference for MetaTables.
-The runnable beginner walkthrough is maintained in the separate canonical
-tutorial CodeRepository at
-<https://github.com/mainsequence-projects/mainsequence-sdk-tutorial>; read it
-only when a runnable end-to-end example is needed.
+The pages above are the authoritative reference for MetaTables.
 
 ## Inputs This Skill Needs
 
@@ -313,9 +310,9 @@ asset_meta_table = MetaTable.register(asset_request)
 ### 4. Schema changes use Alembic
 
 When doing migration work, use
-`.agents/skills/mainsequence/data_publishing/meta_table_migrations/SKILL.md`
-and read `docs/knowledge/meta_tables/migrations.md`. The migration skill owns the
-provider-based Alembic lifecycle.
+`.agents/skills/mainsequence/data_publishing/meta_table_migrations/SKILL.md`,
+which owns the provider-based Alembic lifecycle and carries the migration
+reading list.
 
 Do not apply in-place contract changes by changing a `PlatformManagedMetaTable`
 SQLAlchemy class and calling normal registration again. Shape-addressed
@@ -355,8 +352,8 @@ physical table and Alembic version table names. Use `suffix` for a namespace or
 variant, for example `schema_table_name("msm", "positions", suffix="broker")`.
 
 Do not ask users to construct backend migration payloads or call low-level
-migration request models. The backend request shape is reference material in
-the tutorial; the user-facing path is:
+migration request models. The backend request shape is internal; the
+user-facing path is:
 
 ```bash
 mainsequence migrations current --provider sdk_examples.migrations:migration

{mainsequence-8.1.20.dev8 → mainsequence-8.1.20.dev9}/agent_scaffold/skills/data_publishing/time_index_table_updates/SKILL.md

@@ -64,8 +64,9 @@ If the task depends on one of those areas, route it explicitly instead of guessi
 
 ## Read First
 
-1. `docs/knowledge/time_index_table_updates.md`
-2. `docs/knowledge/meta_tables/sqlalchemy.md`
+1. `AGENTS.md`
+2. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/time_index_table_updates/>
+3. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/meta_tables/sqlalchemy/>
 
 ## Inputs This Skill Needs
 

{mainsequence-8.1.20.dev8 → mainsequence-8.1.20.dev9}/agent_scaffold/skills/platform_operations/access_control_and_sharing/SKILL.md

@@ -61,16 +61,13 @@ This skill only reasons about access to deployed resources such as `ResourceRele
 
 ## Read First
 
-1. `docs/knowledge/infrastructure/users_and_access.md`
-2. `docs/knowledge/infrastructure/constants_and_secrets.md`
+1. `AGENTS.md`
+2. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/infrastructure/users_and_access/>
+3. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/infrastructure/constants_and_secrets/>
 
-If the task is specifically about a resource type, also read the corresponding knowledge or CLI page for that resource.
-
-The knowledge pages above are the authoritative SDK reference for access control
-and sharing. The runnable beginner walkthrough is maintained in the separate
-canonical tutorial CodeRepository at
-<https://github.com/mainsequence-projects/mainsequence-sdk-tutorial>; read it
-only when a runnable end-to-end example is needed.
+The pages above are the authoritative reference for access control and sharing.
+If the task is specifically about a resource type, also read that resource's own
+page from the documentation root at <https://mainsequence-sdk.github.io/mainsequence-sdk/>.
 
 ## Inputs This Skill Needs
 

{mainsequence-8.1.20.dev8 → mainsequence-8.1.20.dev9}/agent_scaffold/skills/platform_operations/orchestration_and_releases/SKILL.md

@@ -57,10 +57,14 @@ This skill must not claim ownership of:
 
 ## Read First
 
-1. `docs/knowledge/infrastructure/scheduling_jobs.md`
-2. `docs/knowledge/infrastructure/artifacts.md`
-3. `docs/knowledge/infrastructure/owner_observability.md` when inspecting logs or resource usage
-If the task touches deployed FastAPI APIs, also read the relevant API skill/docs before changing the operational workflow.
+1. `AGENTS.md`
+2. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/infrastructure/scheduling_jobs/>
+3. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/infrastructure/artifacts/>
+4. <https://mainsequence-sdk.github.io/mainsequence-sdk/knowledge/infrastructure/owner_observability/> when inspecting logs or resource usage
+
+If the task touches deployed FastAPI APIs, also use
+`.agents/skills/mainsequence/application_surfaces/api_surfaces/SKILL.md` before
+changing the operational workflow.
 
 ## Inputs This Skill Needs
 

{mainsequence-8.1.20.dev8 → mainsequence-8.1.20.dev9/mainsequence.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mainsequence
-Version: 8.1.20.dev8
+Version: 8.1.20.dev9
 Summary: Main Sequence SDK 
 Author-email: Main Sequence GmbH <dev@main-sequence.io>
 License-Expression: LicenseRef-MainSequence-Proprietary

{mainsequence-8.1.20.dev8 → mainsequence-8.1.20.dev9}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "mainsequence"
-version = "8.1.20.dev8"
+version = "8.1.20.dev9"
 description = "Main Sequence SDK "
 readme = "README.md"
 requires-python = ">=3.13"

{mainsequence-8.1.20.dev8 → mainsequence-8.1.20.dev9}/tests/test_code_repository_documentation_contract.py

@@ -24,10 +24,14 @@ STREAMLIT_REMOVAL_LINK_PATHS = {
     Path("docs/index.md"),
 }
 SKILL_ROOT = REPOSITORY_ROOT / "agent_scaffold" / "skills"
-# Skills ship inside the wheel but reference repository documentation by its
-# repository-relative path, so every referenced path must exist in the tree that
-# produced the release.
-SKILL_DOCUMENTATION_REFERENCE = re.compile(r"`(docs/[A-Za-z0-9_./-]+\.md)`")
+DOCUMENTATION_SITE_ROOT = "https://mainsequence-sdk.github.io/mainsequence-sdk"
+# The wheel ships `agent_scaffold/skills` and no `docs/`, and the CLI copies those
+# skills into a user's CodeRepository under `.agents/skills/<namespace>/`. There,
+# `docs/` is that repository's own directory, so a skill that names an SDK
+# documentation page by repository-relative path names a file the agent reading it
+# cannot open. Skills cite the published documentation site instead.
+SKILL_REPOSITORY_DOCUMENTATION_PATH = re.compile(r"`(docs/[A-Za-z0-9_./-]+\.md)`")
+SKILL_DOCUMENTATION_URL = re.compile(re.escape(DOCUMENTATION_SITE_ROOT) + r"(/[A-Za-z0-9_./-]*)")
 # Skills are copied into a CodeRepository under `.agents/skills/<namespace>/`, so
 # a cross-skill reference maps back to a shipped `agent_scaffold/skills` path.
 SKILL_CROSS_REFERENCE = re.compile(r"`\.agents/skills/[A-Za-z0-9_]+/([A-Za-z0-9_./-]+/SKILL\.md)`")
@@ -54,8 +58,33 @@ RETIRED_PUBLIC_IDENTIFIERS = {
 }
 
 
-def _skill_files() -> tuple[Path, ...]:
-    return tuple(sorted(SKILL_ROOT.rglob("SKILL.md")))
+def _shipped_scaffold_documents() -> tuple[Path, ...]:
+    """Every Markdown file the wheel ships under `agent_scaffold`.
+
+    `AGENTS.md` ships beside the skills and several skills make it their first
+    mandatory read, so it carries the same reference obligations they do.
+    """
+
+    return tuple(sorted((REPOSITORY_ROOT / "agent_scaffold").rglob("*.md")))
+
+
+def _documentation_page_for_url_path(url_path: str) -> Path | None:
+    """Resolve a documentation-site URL path to the `docs/` page that renders it.
+
+    MkDocs serves directory URLs, so `/knowledge/meta_tables/migrations/` renders
+    `docs/knowledge/meta_tables/migrations.md` and `/knowledge/meta_tables/`
+    renders that directory's `index.md`. The transform is pure, so a skill's
+    documentation links are checkable without reaching the network.
+    """
+
+    relative = url_path.strip("/")
+    if not relative:
+        return REPOSITORY_ROOT / "docs" / "index.md"
+    candidates = (
+        REPOSITORY_ROOT / "docs" / f"{relative}.md",
+        REPOSITORY_ROOT / "docs" / relative / "index.md",
+    )
+    return next((candidate for candidate in candidates if candidate.is_file()), None)
 
 
 def _documentation_files() -> tuple[Path, ...]:
@@ -134,27 +163,48 @@ def test_current_documentation_does_not_advertise_streamlit_support(path: Path)
 
 @pytest.mark.parametrize(
     "path",
-    _skill_files(),
+    _shipped_scaffold_documents(),
     ids=lambda path: str(path.relative_to(REPOSITORY_ROOT)),
 )
-def test_skill_documentation_references_resolve(path: Path) -> None:
+def test_skills_do_not_cite_repository_relative_documentation_paths(path: Path) -> None:
     relative_path = path.relative_to(REPOSITORY_ROOT)
     text = path.read_text(encoding="utf-8")
 
-    missing = sorted(
+    repository_paths = sorted(set(SKILL_REPOSITORY_DOCUMENTATION_PATH.findall(text)))
+
+    assert not repository_paths, (
+        f"{relative_path} cites SDK documentation by repository-relative path: "
+        f"{repository_paths}. The wheel ships no `docs/`, so cite "
+        f"{DOCUMENTATION_SITE_ROOT}/... instead."
+    )
+
+
+@pytest.mark.parametrize(
+    "path",
+    _shipped_scaffold_documents(),
+    ids=lambda path: str(path.relative_to(REPOSITORY_ROOT)),
+)
+def test_skill_documentation_urls_render_an_existing_page(path: Path) -> None:
+    relative_path = path.relative_to(REPOSITORY_ROOT)
+    text = path.read_text(encoding="utf-8")
+
+    unrendered = sorted(
         {
-            reference
-            for reference in SKILL_DOCUMENTATION_REFERENCE.findall(text)
-            if not (REPOSITORY_ROOT / reference).is_file()
+            url_path
+            for url_path in SKILL_DOCUMENTATION_URL.findall(text)
+            if _documentation_page_for_url_path(url_path) is None
         }
     )
 
-    assert not missing, f"{relative_path} references documentation that does not exist: {missing}"
+    assert not unrendered, (
+        f"{relative_path} links documentation-site paths that no `docs/` page "
+        f"renders: {unrendered}"
+    )
 
 
 @pytest.mark.parametrize(
     "path",
-    _skill_files(),
+    _shipped_scaffold_documents(),
     ids=lambda path: str(path.relative_to(REPOSITORY_ROOT)),
 )
 def test_skill_cross_references_resolve(path: Path) -> None: