ai-docs-toolkit 0.4.2__tar.gz → 0.5.0__tar.gz

{ai_docs_toolkit-0.4.2/src/ai_docs_toolkit.egg-info → ai_docs_toolkit-0.5.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ai-docs-toolkit
-Version: 0.4.2
+Version: 0.5.0
 Summary: AI-oriented documentation toolkit.
 License-Expression: MIT
 Requires-Python: >=3.14
@@ -33,7 +33,7 @@ The core idea is that documentation should not be a secondary artifact created a
 
 ## Current Status
 
-Version 0.4.2 includes:
+Version 0.5.0 includes:
 
 - schema and structure validation for Markdown documents with YAML front matter;
 - document graph output;
@@ -45,32 +45,62 @@ Version 0.4.2 includes:
 - context bundle output;
 - an executable `ai-docs` CLI;
 - optional local MCP runtime entrypoint.
+- opt-in workspace impact analysis through CLI and MCP, with explicit manifests,
+  repository-scoped IDs, public event interfaces and visible partial results.
 
 ## Installation
 
-Install version 0.4.2 with Python 3.14 or newer:
+Install version 0.5.0 with Python 3.14 or newer:
 
 ```bash
-python -m pip install ai-docs-toolkit==0.4.2
+python -m pip install ai-docs-toolkit==0.5.0
 ai-docs --version
 ```
 
 For ephemeral execution with `uvx`:
 
 ```bash
-uvx --python 3.14 --from ai-docs-toolkit==0.4.2 ai-docs --version
+uvx --python 3.14 --from ai-docs-toolkit==0.5.0 ai-docs --version
 ```
 
 For the optional local MCP runtime:
 
 ```bash
-python -m pip install "ai-docs-toolkit[mcp]==0.4.2"
+python -m pip install "ai-docs-toolkit[mcp]==0.5.0"
 ai-docs-mcp --project-root .
 ```
 
 For the latest stable release, use `python -m pip install --upgrade ai-docs-toolkit`
 without an exact version pin.
 
+## Release 0.5.0
+
+`ai-docs workspace impact --workspace ./ai-docs.workspace.yaml --input ./change.yaml --format json`
+analyzes explicitly connected independent repositories. Stable repository IDs
+namespace document IDs; a system repository owns public event contracts and
+producer/consumer roles, while local bindings select implementation modules.
+Modules may be nested within a repository. Cross-repository `related.*` links
+and inferred dependency discovery are not required.
+
+The MCP tool is registered only when the server starts with `--workspace`;
+ordinary local commands retain their scope. JSON v1 preserves candidates,
+provenance and diagnostics when a checkout, binding or interface version is
+missing. A partial result is not a clean review, and resolved routing does not
+prove sufficient context or implementation compatibility.
+
+Upgrading from 0.4.2 refreshes four managed workspace schemas and the consumer
+reference in both `minimal` and `agent` profiles. Review `init --dry-run` and
+`init --check` with the existing profile, reconcile local modifications, then
+run ordinary `init` without blanket `--force`. Project documents, IDs, local
+policy and configuration remain project-owned. Repository IDs, manifests and
+bindings require manual opt-in; init does not discover neighboring repositories.
+Exact commands and rollback are in “Migration 0.4.2 To 0.5.0” in the packaged
+`ai_docs_toolkit/docs/agent-installation.md`.
+
+Verification includes installed CLI/MCP and upgrade matrices plus a bounded
+HAM documentation evaluation. Its accepted result retains native-format
+validation findings; it is not a full HAM migration or service runtime check.
+
 ## Release 0.4.2
 
 The package now declares the MIT license in its metadata and includes the full

{ai_docs_toolkit-0.4.2 → ai_docs_toolkit-0.5.0}/README.md

@@ -2,7 +2,7 @@
 
 AI Docs Toolkit is a documentation toolkit for AI-assisted software development.
 
-It treats documentation as an engineering interface between humans, AI agents, code, tests and change management. The 0.4.2 source declares the MIT license and retains the 0.4.1 impact, review and code-link improvements. Python 3.14 or newer is required.
+It treats documentation as an engineering interface between humans, AI agents, code, tests and change management. Version 0.5.0 adds opt-in workspace impact analysis across independent repositories through CLI and MCP. The package uses the MIT license and requires Python 3.14 or newer.
 
 ## Purpose
 
@@ -34,6 +34,7 @@ The stable package provides:
 - impact analysis;
 - context selection with history filtering, optional body budgets and configurable depth;
 - proportionate agent workflows and optional MCP integration;
+- explicit workspace manifests, repository-scoped document IDs and public event interface routing;
 - CI, pre-commit and agent workflow examples.
 
 At this stage the repository contains:
@@ -83,10 +84,13 @@ pipx install ai-docs-toolkit
 uvx --from ai-docs-toolkit ai-docs validate
 ```
 
-[Version 0.4.1 is available on Public PyPI](https://pypi.org/project/ai-docs-toolkit/0.4.1/).
-Version 0.4.2 is prepared for publication with MIT license metadata and updated
-product guidance. It changes no CLI, schema, template or runtime behavior.
-See the [0.4.1 → 0.4.2 upgrade note](src/ai_docs_toolkit/docs/agent-installation.md#migration-041-to-042).
+[Release 0.5.0](ai-docs/reference/releases/0-5-0.md) adds `ai-docs workspace impact`
+and the explicitly enabled MCP workspace tool. Existing repository-local commands
+keep their scope. See the [0.4.2 → 0.5.0 migration guide](src/ai_docs_toolkit/docs/agent-installation.md#migration-042-to-050)
+for the managed schema/reference refresh in both init profiles and manual opt-in.
+Registry availability is shown on [Public PyPI](https://pypi.org/project/ai-docs-toolkit/).
+If upgrading from 0.4.1, also read the
+[0.4.1 → 0.4.2 upgrade note](src/ai_docs_toolkit/docs/agent-installation.md#migration-041-to-042).
 If upgrading from 0.4.0, also follow the
 [0.4.0 → 0.4.1 migration guide](src/ai_docs_toolkit/docs/agent-installation.md#migration-040-to-041).
 That patch required no schema, configuration or managed-template content migration;
@@ -270,7 +274,7 @@ The MVP does not aim to provide:
 - full IDE plugins;
 - hosted services;
 - database-backed graph storage;
-- cross-repository dependency graphs.
+- inferred or transitive cross-repository dependency graphs beyond explicit public interface bindings.
 
 ## License
 

{ai_docs_toolkit-0.4.2 → ai_docs_toolkit-0.5.0}/packaging/pypi/README.md

@@ -16,7 +16,7 @@ The core idea is that documentation should not be a secondary artifact created a
 
 ## Current Status
 
-Version 0.4.2 includes:
+Version 0.5.0 includes:
 
 - schema and structure validation for Markdown documents with YAML front matter;
 - document graph output;
@@ -28,32 +28,62 @@ Version 0.4.2 includes:
 - context bundle output;
 - an executable `ai-docs` CLI;
 - optional local MCP runtime entrypoint.
+- opt-in workspace impact analysis through CLI and MCP, with explicit manifests,
+  repository-scoped IDs, public event interfaces and visible partial results.
 
 ## Installation
 
-Install version 0.4.2 with Python 3.14 or newer:
+Install version 0.5.0 with Python 3.14 or newer:
 
 ```bash
-python -m pip install ai-docs-toolkit==0.4.2
+python -m pip install ai-docs-toolkit==0.5.0
 ai-docs --version
 ```
 
 For ephemeral execution with `uvx`:
 
 ```bash
-uvx --python 3.14 --from ai-docs-toolkit==0.4.2 ai-docs --version
+uvx --python 3.14 --from ai-docs-toolkit==0.5.0 ai-docs --version
 ```
 
 For the optional local MCP runtime:
 
 ```bash
-python -m pip install "ai-docs-toolkit[mcp]==0.4.2"
+python -m pip install "ai-docs-toolkit[mcp]==0.5.0"
 ai-docs-mcp --project-root .
 ```
 
 For the latest stable release, use `python -m pip install --upgrade ai-docs-toolkit`
 without an exact version pin.
 
+## Release 0.5.0
+
+`ai-docs workspace impact --workspace ./ai-docs.workspace.yaml --input ./change.yaml --format json`
+analyzes explicitly connected independent repositories. Stable repository IDs
+namespace document IDs; a system repository owns public event contracts and
+producer/consumer roles, while local bindings select implementation modules.
+Modules may be nested within a repository. Cross-repository `related.*` links
+and inferred dependency discovery are not required.
+
+The MCP tool is registered only when the server starts with `--workspace`;
+ordinary local commands retain their scope. JSON v1 preserves candidates,
+provenance and diagnostics when a checkout, binding or interface version is
+missing. A partial result is not a clean review, and resolved routing does not
+prove sufficient context or implementation compatibility.
+
+Upgrading from 0.4.2 refreshes four managed workspace schemas and the consumer
+reference in both `minimal` and `agent` profiles. Review `init --dry-run` and
+`init --check` with the existing profile, reconcile local modifications, then
+run ordinary `init` without blanket `--force`. Project documents, IDs, local
+policy and configuration remain project-owned. Repository IDs, manifests and
+bindings require manual opt-in; init does not discover neighboring repositories.
+Exact commands and rollback are in “Migration 0.4.2 To 0.5.0” in the packaged
+`ai_docs_toolkit/docs/agent-installation.md`.
+
+Verification includes installed CLI/MCP and upgrade matrices plus a bounded
+HAM documentation evaluation. Its accepted result retains native-format
+validation findings; it is not a full HAM migration or service runtime check.
+
 ## Release 0.4.2
 
 The package now declares the MIT license in its metadata and includes the full

{ai_docs_toolkit-0.4.2 → ai_docs_toolkit-0.5.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "ai-docs-toolkit"
-version = "0.4.2"
+version = "0.5.0"
 description = "AI-oriented documentation toolkit."
 readme = "packaging/pypi/README.md"
 requires-python = ">=3.14"
@@ -41,6 +41,7 @@ ai_docs_toolkit = [
   "templates/project/.ai-docs/reference/*.md",
   "templates/project/.ai-docs/schemas/*.schema.yaml",
   "templates/project/.ai-docs/schemas/document-types/*.schema.yaml",
+  "templates/project/.ai-docs/schemas/workspace/*-v1.schema.yaml",
   "templates/project/.ai-docs/templates/release/*.md",
   "templates/project/.ai-docs/workflows/*.md",
 ]

{ai_docs_toolkit-0.4.2 → ai_docs_toolkit-0.5.0}/src/ai_docs_toolkit/__init__.py

@@ -1,3 +1,3 @@
 """AI Docs Toolkit package baseline."""
 
-__version__ = "0.4.2"
+__version__ = "0.5.0"

{ai_docs_toolkit-0.4.2 → ai_docs_toolkit-0.5.0}/src/ai_docs_toolkit/cli/__init__.py

@@ -91,6 +91,7 @@ def _help_text() -> str:
             "  ai-docs tasks push --provider azure-devops (--task TASK-ID | --stage STAGE-ID | --plan PLAN-ID) --organization ORG --project PROJECT [--dry-run] [--write-back] [--format json]",
             "  ai-docs graph --format json [--type TYPE] [--module MODULE] [--id ID]",
             "  ai-docs impact (--id ID | --changed [--base REF]) [--format json]",
+            "  ai-docs workspace impact --workspace FILE --input FILE [--format human|json]",
             "  ai-docs code-links-check FILE... [--format human|json|--json]",
             "  ai-docs review-check RECORD [--record N] [--format json|--json]",
             "  ai-docs context (--id ID | --module MODULE | --feature FEATURE | --changed) [--format json] [--history exclude|include|legacy] [--body-budget-bytes N] [--depth N]",
@@ -110,6 +111,11 @@ def main(argv: list[str] | None = None) -> int:
 
         args = sys.argv[1:]
 
+    if args and args[0] == "workspace":
+        from .workspace import main as workspace_main
+
+        return workspace_main(args[1:])
+
     if not args or "--help" in args or "-h" in args:
         print(_help_text())
         return 0

ai_docs_toolkit-0.5.0/src/ai_docs_toolkit/cli/workspace.py

@@ -0,0 +1,58 @@
+"""Explicit workspace CLI adapter; domain results come from the shared service."""
+
+import argparse
+import sys
+
+from ai_docs_toolkit.workspace import analyze_workspace_file, format_workspace_json
+
+
+class UsageError(ValueError):
+    pass
+
+
+class WorkspaceParser(argparse.ArgumentParser):
+    def error(self, message):
+        raise UsageError(message)
+
+
+def main(argv: list[str]) -> int:
+    parser = WorkspaceParser(prog="ai-docs workspace impact", allow_abbrev=False)
+    parser.add_argument("--workspace", required=True, help="Explicit workspace manifest path.")
+    parser.add_argument("--input", required=True, help="Saved JSON/YAML request v1 path.")
+    parser.add_argument("--format", choices=("human", "json"), default="human")
+    try:
+        if not argv or argv[0] != "impact":
+            raise UsageError("expected 'workspace impact'")
+        # Duplicate selectors are ambiguous even if argparse would accept the last value.
+        options = [arg.split("=", 1)[0] for arg in argv[1:] if arg.startswith("--")]
+        if len(options) != len(set(options)):
+            raise UsageError("duplicate option")
+        args = parser.parse_args(argv[1:])
+    except UsageError as exc:
+        parser.print_usage(sys.stderr)
+        print(f"{parser.prog}: error: {exc}", file=sys.stderr)
+        return 3
+    except SystemExit as exc:
+        return int(exc.code)
+    result = analyze_workspace_file(args.workspace, args.input)
+    print(format_workspace_json(result) if args.format == "json" else format_human(result), end="")
+    return result["exit_code"]
+
+
+def format_human(result: dict) -> str:
+    lines = [f"Workspace routing: {result['routing_status']} (exit {result['exit_code']})",
+             f"Sources: {len(result['sources'])}; candidates: {len(result['candidates'])}"]
+    for kind in ("sources", "candidates"):
+        for document in result[kind]:
+            lines.append(f"{kind}: {document['repository_id']}:{document['document_id']}")
+    for gap in result["gaps"]:
+        lines.append(f"Gap {gap['code']} [{gap['repository_id']}]: {gap['message']}")
+    for review in result["reviews"]:
+        if review["system_review_required"]:
+            lines.append(f"System review required: {review['repository_id']} ({review['interface_impact']})")
+    for context in result["context"]:
+        for field in ("warnings", "gaps", "excluded", "historical"):
+            if context.get(field):
+                lines.append(f"Context {context['repository_id']} {field}: {len(context[field])}")
+    lines.append("Routing completion does not mean review completion or sufficient context. Use --format json for evidence.")
+    return "\n".join(lines) + "\n"

{ai_docs_toolkit-0.4.2 → ai_docs_toolkit-0.5.0}/src/ai_docs_toolkit/core/changed_files.py

@@ -49,9 +49,15 @@ def read_git_changed_files(
     *,
     unavailable_rule_id: str,
     base_ref: str | None = None,
+    git_config: tuple[str, ...] = (),
 ) -> GitChangedFilesResult:
+    """Discover local paths; optional process config lets isolated callers disable hooks.
+
+    Workspace analysis passes `-c core.fsmonitor=false` so reading a checkout
+    cannot execute its monitor program. Existing callers retain their defaults.
+    """
     process = subprocess.run(
-        ["git", "status", "--porcelain", "-z", "--untracked-files=all"],
+        ["git", *git_config, "status", "--porcelain", "-z", "--untracked-files=all"],
         cwd=root,
         capture_output=True,
         text=True,
@@ -76,6 +82,7 @@ def read_git_changed_files(
         diff_process = subprocess.run(
             [
                 "git",
+                *git_config,
                 "diff",
                 "--name-only",
                 "-z",

{ai_docs_toolkit-0.4.2 → ai_docs_toolkit-0.5.0}/src/ai_docs_toolkit/core/init.py

@@ -440,6 +440,36 @@ BOOTSTRAP_TEMPLATES = (
         profiles=frozenset({"agent"}),
         overwrite_existing=True,
     ),
+    BootstrapTemplate(
+        source_name=".ai-docs/schemas/workspace/manifest-v1.schema.yaml",
+        target_path=".ai-docs/schemas/workspace/manifest-v1.schema.yaml",
+        profiles=frozenset({"minimal", "agent"}),
+        overwrite_existing=True,
+    ),
+    BootstrapTemplate(
+        source_name=".ai-docs/schemas/workspace/request-v1.schema.yaml",
+        target_path=".ai-docs/schemas/workspace/request-v1.schema.yaml",
+        profiles=frozenset({"minimal", "agent"}),
+        overwrite_existing=True,
+    ),
+    BootstrapTemplate(
+        source_name=".ai-docs/schemas/workspace/repository-v1.schema.yaml",
+        target_path=".ai-docs/schemas/workspace/repository-v1.schema.yaml",
+        profiles=frozenset({"minimal", "agent"}),
+        overwrite_existing=True,
+    ),
+    BootstrapTemplate(
+        source_name=".ai-docs/schemas/workspace/result-v1.schema.yaml",
+        target_path=".ai-docs/schemas/workspace/result-v1.schema.yaml",
+        profiles=frozenset({"minimal", "agent"}),
+        overwrite_existing=True,
+    ),
+    BootstrapTemplate(
+        source_name=".ai-docs/reference/workspace-analysis.md",
+        target_path=".ai-docs/reference/workspace-analysis.md",
+        profiles=frozenset({"minimal", "agent"}),
+        overwrite_existing=True,
+    ),
 )
 
 

{ai_docs_toolkit-0.4.2 → ai_docs_toolkit-0.5.0}/src/ai_docs_toolkit/docs/agent-installation.md

@@ -47,7 +47,7 @@ ai-docs --version
 Pinned install:
 
 ```bash
-python -m pip install "ai-docs-toolkit==0.4.2"
+python -m pip install "ai-docs-toolkit==0.5.0"
 ai-docs --version
 ```
 
@@ -89,6 +89,96 @@ bulk task deletion is required. Detailed migration and rollback guidance is in
 the installed `.ai-docs/reference/planning-document-model.md`, under
 “Consumer Migration: Proportionate Workflow”.
 
+## Migration 0.4.2 To 0.5.0
+
+From: `0.4.2`. To: `0.5.0`. Python 3.14+ remains required. This minor release
+adds opt-in workspace impact analysis through CLI/MCP and four workspace v1
+schemas. Ordinary local graph, impact and context commands keep their scope.
+No product-document or service-runtime migration is performed by the package.
+
+Before upgrading, record the package version, Git state, hashes of project-owned
+files and the results of independent project checks. Preserve saved review rounds:
+
+```bash
+ai-docs --version
+ai-docs validate --json
+ai-docs doctor --format json
+# Use the project's existing profile, either agent or minimal:
+ai-docs init --profile agent --dry-run
+ai-docs init --profile agent --check
+```
+
+After the version is available on Public PyPI, upgrade in the approved environment
+and update the project's version pin. Choose the applicable installation method:
+
+```bash
+python -m pip install --upgrade "ai-docs-toolkit==0.5.0"
+# MCP users install the complete extra instead:
+python -m pip install --upgrade "ai-docs-toolkit[mcp]==0.5.0"
+pipx install --force --python python3.14 "ai-docs-toolkit==0.5.0"
+uvx --python 3.14 --from ai-docs-toolkit==0.5.0 ai-docs --version
+```
+
+Reconnect long-lived MCP clients. Inspect `init --dry-run` and `--check` again
+using the same profile. Both profiles add `.ai-docs/schemas/workspace/` with
+`repository-v1.schema.yaml`, `manifest-v1.schema.yaml`, `request-v1.schema.yaml`
+and `result-v1.schema.yaml`, plus `.ai-docs/reference/workspace-analysis.md`.
+Known unchanged managed files can be refreshed by ordinary init:
+
+```bash
+ai-docs init --profile agent --dry-run
+ai-docs init --profile agent --check
+# After reviewing the proposed managed-file changes:
+ai-docs init --profile agent
+ai-docs init --profile agent --check
+ai-docs --version
+ai-docs validate --json
+ai-docs doctor --format json
+```
+
+Substitute `--profile minimal` throughout for an existing minimal project.
+`locally_modified` or `provenance_missing` blocks ordinary init before any
+upgrade writes. Reconcile those files and their provenance manually before
+retrying; do not use blanket `--force` as a migration shortcut. An initial
+nonzero `init --check` can describe the expected pending refresh; after applying
+reviewed updates it must be clean and a repeat init must be idempotent.
+
+Preserve `ai-docs.config.yaml`, existing AGENTS.md and index documents,
+`.ai-docs/local/**`, product documents, IDs, aliases, links, task/review history
+and project-owned workspace manifests. Managed provenance is recorded in
+`.ai-docs/managed-files.json`. Compare project-owned hashes after the upgrade;
+stop for unexpected rewrites or new unexplained validation/doctor findings.
+Keep all independent project checks. A toolkit update does not make an
+unconfigured or partially adapted documentation corpus valid automatically.
+
+Workspace use is optional. To enable it, follow the installed
+`.ai-docs/reference/workspace-analysis.md`: manually assign stable repository
+IDs, configure the system public event interfaces and local module bindings,
+then create a project-owned manifest and request. Init does not create them.
+Use explicit CLI paths, or add startup `--workspace` to the MCP server:
+
+```bash
+ai-docs workspace impact --workspace ./ai-docs.workspace.yaml --input ./change.yaml --format json
+ai-docs-mcp --project-root . --workspace ./ai-docs.workspace.yaml
+```
+
+Verify the selected candidates and review context against the intended change.
+Exit 1/`partial` retains available candidates and reports gaps; it is not zero
+impact. Exit 0/`resolved` does not complete semantic review or run service tests.
+MCP exposes the workspace tool only after explicit server startup opt-in;
+clients cannot override the manifest path per call. Separate local commands
+remain usable without neighboring checkouts.
+
+To disable workspace analysis, remove the workspace invocation from automation
+and its startup argument from MCP, then restart the server. For rollback,
+preserve the current evidence, restore the previous approved pin and reinstall
+`ai-docs-toolkit==0.4.2` (or the complete MCP extra) in the supported environment.
+Restore only this upgrade's reviewed managed-file/provenance changes from Git,
+preserving unrelated project work, and repeat local checks. Version 0.4.2 has
+no workspace command. Package rollback does not rewrite project files or undo
+manual opt-in metadata; retain that metadata and history unless separately
+authorized to change them.
+
 ## Migration 0.4.1 To 0.4.2
 
 From: `0.4.1`. To: `0.4.2`. Install the target after it appears on Public

{ai_docs_toolkit-0.4.2 → ai_docs_toolkit-0.5.0}/src/ai_docs_toolkit/mcp/server.py

@@ -30,6 +30,7 @@ SERVER_NAME = "AI Docs Toolkit"
 class McpRuntimeArgs:
     project_root: Path
     config_file: str
+    workspace: str | None = None
 
 
 def parse_args(argv: list[str] | None = None) -> McpRuntimeArgs:
@@ -47,10 +48,12 @@ def parse_args(argv: list[str] | None = None) -> McpRuntimeArgs:
         default=DEFAULT_CONFIG_FILE,
         help=f"Project config file relative to the project root. Defaults to {DEFAULT_CONFIG_FILE}.",
     )
+    parser.add_argument("--workspace", help="Opt in to a workspace manifest, relative to project root.")
     parsed = parser.parse_args(argv)
     return McpRuntimeArgs(
         project_root=Path(parsed.project_root),
         config_file=parsed.config,
+        workspace=parsed.workspace,
     )
 
 
@@ -186,15 +189,33 @@ def build_server(
     *,
     project_root: str | Path = ".",
     config_file: str = DEFAULT_CONFIG_FILE,
+    workspace: str | Path | None = None,
     mcp_server_factory: Callable[..., Any] | None = None,
 ) -> Any:
     server_factory = mcp_server_factory or _load_mcp_server()
     server = server_factory(SERVER_NAME, version=__version__)
     default_project_root = project_root
+    # Capture an absolute startup choice; calls cannot override it or change it via cwd.
+    workspace_path = (Path(project_root) / workspace).resolve() if workspace is not None else None
+
+    def runtime_info() -> dict[str, Any]:
+        info = server_info(project_root=project_root, config_file=config_file)
+        if workspace_path is not None:
+            info["workspace"] = {"path": str(workspace_path),
+                                 "request_schema_versions": [1], "result_schema_versions": [1]}
+        return info
 
     @server.tool(name="ai_docs_server_info")
     def ai_docs_server_info() -> dict[str, Any]:
-        return server_info(project_root=project_root, config_file=config_file)
+        return runtime_info()
+
+    if workspace_path is not None:
+        from ai_docs_toolkit.workspace import analyze_workspace
+
+        @server.tool(name="ai_docs_workspace_impact")
+        def ai_docs_workspace_impact(request: dict[str, Any]) -> dict[str, Any]:
+            """Analyze request v1 against the manifest selected at server startup."""
+            return analyze_workspace(workspace_path, request)
 
     @server.tool(name="ai_docs_validate")
     def ai_docs_validate(project_root: str | None = None) -> dict[str, Any]:
@@ -248,7 +269,7 @@ def build_server(
 
     @server.resource("ai-docs://config")
     def ai_docs_config() -> dict[str, Any]:
-        return server_info(project_root=project_root, config_file=config_file)
+        return runtime_info()
 
     @server.resource("ai-docs://graph")
     def ai_docs_graph_resource() -> dict[str, Any]:
@@ -283,7 +304,7 @@ def build_server(
 
 def main(argv: list[str] | None = None) -> int:
     args = parse_args(argv)
-    server = build_server(project_root=args.project_root, config_file=args.config_file)
+    server = build_server(project_root=args.project_root, config_file=args.config_file, workspace=args.workspace)
     server.run()
     return 0
 

{ai_docs_toolkit-0.4.2 → ai_docs_toolkit-0.5.0}/src/ai_docs_toolkit/templates/project/.ai-docs/reference/init-command-contract.md

@@ -114,6 +114,11 @@ MVP target paths are repository-root relative:
 | `.ai-docs/schemas/document-types/runbook.schema.yaml` | `minimal`, `agent` | Create or overwrite from toolkit distribution. |
 | `.ai-docs/schemas/document-types/scenario.schema.yaml` | `minimal`, `agent` | Create or overwrite from toolkit distribution. |
 | `.ai-docs/schemas/document-types/test-case.schema.yaml` | `minimal`, `agent` | Create or overwrite from toolkit distribution. |
+| `.ai-docs/schemas/workspace/manifest-v1.schema.yaml` | `minimal`, `agent` | Managed schema v1; existing provenance/preflight rules apply. |
+| `.ai-docs/schemas/workspace/request-v1.schema.yaml` | `minimal`, `agent` | Managed schema v1; existing provenance/preflight rules apply. |
+| `.ai-docs/schemas/workspace/repository-v1.schema.yaml` | `minimal`, `agent` | Managed schema v1; existing provenance/preflight rules apply. |
+| `.ai-docs/schemas/workspace/result-v1.schema.yaml` | `minimal`, `agent` | Managed schema v1; existing provenance/preflight rules apply. |
+| `.ai-docs/reference/workspace-analysis.md` | `minimal`, `agent` | Managed opt-in and upgrade guidance; never creates a manifest. |
 | `.ai-docs/skills/index.md` | `agent` | Create or overwrite from toolkit distribution during planning workflow migration. |
 | `.ai-docs/skills/planning-task-execution/SKILL.md` | `agent` | Create or overwrite from toolkit distribution during planning workflow migration. |
 | `.ai-docs/skills/planning-record-generation/SKILL.md` | `agent` | Create or overwrite from toolkit distribution during planning workflow migration. |
@@ -242,6 +247,17 @@ Required assets must be available from the installed distribution.
 
 ## Validation Contract
 
+### Workspace opt-in в source
+
+[Opt-in contract and upgrade guidance](workspace-analysis.md) describe the optional extension. Существующие профили `minimal`/`agent` сохраняются. `init`
+не назначает repository ID, не ищет соседей и не создаёт workspace manifest.
+Project-owned config, workspace, AGENTS/index/local policy и docs сохраняются;
+managed-file preflight/provenance не ослабляются. Четыре schemas v1 и
+[opt-in/migration guidance](workspace-analysis.md) включены в оба профиля
+как managed artifacts. Manifest не является managed target. Runtime берёт
+схемы из пакета, а не из изменяемой consumer-копии. Опубликованный 0.4.2
+этого расширения не содержит.
+
 Downstream implementation and test tasks must verify:
 - empty repository bootstrap for `minimal`;
 - empty repository bootstrap for `agent`;