mammoth-cli 2.0.41__tar.gz → 2.0.43__tar.gz

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mammoth-cli
-Version: 2.0.41
+Version: 2.0.43
 Summary: Command-line interface for the Mammoth Analytics platform
 Project-URL: homepage, https://mammoth.io
 Project-URL: repository, https://github.com/EdgeMetric/mammothsdk

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/__init__.py

@@ -6,7 +6,7 @@ not implement a second HTTP client and does not call private SDK members.
 
 from __future__ import annotations
 
-__version__ = "2.0.41"
+__version__ = "2.0.43"
 
 # The versioned machine output envelope contract.
 SCHEMA_VERSION = 1

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/bundled_skill/mammoth-cli/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: mammoth-cli
-version: 2.0.41
+version: 2.0.43
 description: "Use Mammoth Analytics from a terminal: install or authenticate the CLI, discover its live command contract, and safely manage projects, data, views, pipelines (join, merge, pivot, filter, clean), dashboards, exports, and handoffs."
 ---
 
@@ -31,7 +31,9 @@ mammoth project ensure 'PROJECT NAME'   # get-or-create; becomes the active proj
 ```
 
 `doctor` failing is a precondition failure, not a reason to try the business
-command anyway. Production is the `app` endpoint; use `release` only when the
+command anyway. When the connection check fails with a 502, 504 or timeout,
+`mammoth doctor --input '{"wait": 300}'` probes again for up to that many
+seconds; go on only when it passes. Production is the `app` endpoint; use `release` only when the
 task names it, and check `meta.profile`/`auth status` `endpoint` match the
 intended environment before doing anything. If no profile has credentials,
 tell the operator to run `mammoth auth login` in their own terminal (add
@@ -95,9 +97,10 @@ dashboard from t_a and t_b"). Work it out from the data before you build:
 
 1. Upload every file. Each upload result carries a `view` preview of what
    Mammoth made of the file (`view_id`, column types, sample rows,
-   `column_warnings`, and `before_dashboard`). Read every view in Mammoth, not
-   the local files: `view data get VIEW_ID` for each one. The warnings list
-   numbers or dates stored as text, and blanks, each with the fix.
+   `column_warnings`, and `before_dashboard`). Do not open the local files
+   (`cat`, `head`): what counts is what Mammoth made of them. Run `view data
+   get VIEW_ID` for each view. The warnings list numbers or dates stored as
+   text, and blanks, each with the fix.
 2. Find the keys. A column in one view whose values appear in a column of
    the other (`customer_id` and `id`, `order_ref` and `order_no`) is a
    foreign key; the view with many rows per key is the main one.
@@ -172,6 +175,10 @@ your report. If no column links the files, ask before you combine them.
   "limit": 50}`.
 - A timeout, exit 7 or interruption does not prove failure; reconcile an
   `outcome_unknown` (`job get`, `view task list`) before replaying.
+- Before you report, run `mammoth project check PROJECT_ID`. Its `to_report`
+  has one line for each open finding in every view and dashboard (blanks,
+  text numbers, money not shown, columns a dashboard cannot see). Fix each
+  one, or give it one line in the report.
 - Run the [report checklist](references/report-checklist.md) before stating
   a number or calling a step done. When files here disagree,
   [capabilities](references/capabilities.md) wins over a recipe, and a recipe

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/bundled_skill/mammoth-cli/references/capabilities.md

@@ -1,7 +1,7 @@
 # What is proven on release
 
 Generated from `docs/release-capability-matrix.json`; do not edit by hand.
-The CLI publishes 556 commands. 455 of them bind one of the 528 API operations in the matrix; the remainder are local commands (`schema`, `auth`, `doctor`, `log`, ...) or typed variants that share an operation (`view transform *` commands submit through `view.task.add`, except `rename-columns` and `sort`, which set view display properties through `view.update`). 262 bound commands ran once successfully on release, 2 are not supported there, and the rest are untried. Untried is not broken: discover the contract with `mammoth schema get COMMAND_ID`, run it, and treat the structured error envelope as the answer.
+The CLI publishes 557 commands. 455 of them bind one of the 528 API operations in the matrix; the remainder are local commands (`schema`, `auth`, `doctor`, `log`, ...) or typed variants that share an operation (`view transform *` commands submit through `view.task.add`, except `rename-columns` and `sort`, which set view display properties through `view.update`). 262 bound commands ran once successfully on release, 2 are not supported there, and the rest are untried. Untried is not broken: discover the contract with `mammoth schema get COMMAND_ID`, run it, and treat the structured error envelope as the answer.
 
 Status meanings:
 

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/bundled_skill/mammoth-cli/references/command-index.md

@@ -1,6 +1,6 @@
 # Published CLI command catalog
 
-This generated catalog is a lookup table for every command in the published CLI manifest, currently 556; each entry carries its release status line. It is a command-contract index, not an assertion that every backend/OpenAPI operation has a CLI binding. For an exact local CLI contract, run `mammoth schema get COMMAND_ID`; `mammoth capability list` is an API-binding inventory and can omit typed/local CLI routes. For focused workflows, read the [recipes index](recipes/index.md). Sensitive structured input must come from a private file or pipe; never put secrets in literal argv.
+This generated catalog is a lookup table for every command in the published CLI manifest, currently 557; each entry carries its release status line. It is a command-contract index, not an assertion that every backend/OpenAPI operation has a CLI binding. For an exact local CLI contract, run `mammoth schema get COMMAND_ID`; `mammoth capability list` is an API-binding inventory and can omit typed/local CLI routes. For focused workflows, read the [recipes index](recipes/index.md). Sensitive structured input must come from a private file or pipe; never put secrets in literal argv.
 
 Load only the applicable domain file:
 
@@ -31,7 +31,7 @@ Load only the applicable domain file:
 - [log](commands/log.md) — 2 published commands
 - [notification](commands/notification.md) — 5 published commands
 - [parameter](commands/parameter.md) — 14 published commands
-- [project](commands/project.md) — 19 published commands
+- [project](commands/project.md) — 20 published commands
 - [report](commands/report.md) — 1 published commands
 - [schedule](commands/schedule.md) — 5 published commands
 - [schema](commands/schema.md) — 3 published commands

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/bundled_skill/mammoth-cli/references/commands/project.md

@@ -22,6 +22,16 @@ Result: `ProjectBulkUpdateResult`; mutation `high_impact`, confirmation `confirm
 
 Status on release: ran once on CLI 2.0.15 — re-verification 2026-09-18: exit 0 on release with CLI 2.0.15. Fix held: body accepted as ProjectsPatch naming project_id in every value item; confirmation named the workspace scope (message: 'bulk-update roles in projects 23 of workspace 4', hint 'Pass --con…
 
+### `project.check`
+
+Run: `mammoth project check`. Exact input fields: `mammoth schema get project.check`.
+
+Example: `mammoth project check 123`. Placeholders are illustrative; resolve IDs and input from observed reads.
+
+Result: `ProjectCheckResult`; mutation `read`, confirmation `none`, wait policy `not_async`.
+
+Status on release: untried; no live run recorded.
+
 ### `project.checkpoint.list`
 
 Run: `mammoth project checkpoint list`. Exact input fields: `mammoth schema get project.checkpoint.list`.

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/bundled_skill/mammoth-cli/references/recipes/dashboards.md

@@ -99,7 +99,8 @@ mammoth view transform math VIEW_ID --project PROJECT_ID \
   are in your report.
 - `create-blank`, `canvas save` and `pages add` return `deliverable_check`:
   `money_not_shown` (with the `math` command in `fix`), `unit_price_summed`,
-  and one `blank_values` entry for each column on the board that has blanks.
+  `columns_not_on_dashboard` (view columns added after the dashboard was
+  made, with the `create-blank` command in `fix`), and one `blank_values` entry for each column on the board that has blanks.
   Fix each warning, or say in your report why you kept it.
 
 **Read the bindings back, then the numbers.** `dashboard canvas get` returns

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/bundled_skill/mammoth-cli/references/recovery.md

@@ -36,7 +36,7 @@ these commands are not blanket permission to replay a mutation.
 | `4DASH012 DASHBOARD_LEGACY_CREATION_RETIRED` (HTTP 409) from `dashboard create` | legacy engine retired | `dashboard create-blank` or `dashboard v3 generate` |
 | `4GENR001 ... params.data` (HTTP 400) from `dashboard pdf export` | route needs a browser-rendered payload | not available from the CLI; report it, do not retry with `{"data": {}}` |
 | `'destination'` KeyError in a `view export create` job with handler_type csv_file | generic export route is broken for CSV on release | use `view export csv` |
-| `timeout` / `outcome_unknown` (exit 7) | the request may have completed | `job get JOB_ID` if you have one, else re-list the resource (`dataset list`, `view task list VIEW_ID DATASET_ID`) before any replay |
+| `timeout` / `outcome_unknown` (exit 7) | the request may have completed | run the envelope's `recovery_commands` first: `job get JOB_ID`, `project get ID` after a delete (repeat until `resource_not_found`), `dashboard list` after `create-blank` (look for your title), `dashboard canvas get ID` after a canvas or pages write. Else re-list the resource (`dataset list`, `view task list VIEW_ID DATASET_ID`) before any replay |
 | HTTP 502 / connection errors on every command | backend outage | `mammoth doctor`; wait and re-run doctor until it succeeds; do not replay mutations blindly |
 
 Every error envelope from CLI 2.0.18 carries `error.log_ref` `{file, run_id}`.

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/bundled_skill/mammoth-cli/references/report-checklist.md

@@ -8,6 +8,7 @@ Every "yes" must point at a command you ran in this task and its response.
 | State rows in and rows out for every step that can change row count (filter, discard-duplicates, join, limit, unnest, pivot) | `view data get` or `view get` (`row_count`) before and after the step, or the `view data get` around a set-values |
 | Read back every value-changing transform | `view data get VIEW_ID DATASET_ID` (full page; `view data query` with `limit`/`condition` for a slice) after the job reached `success`; the sample shows the intended change and nothing else changed |
 | No measure column is all-zero, all-null, or all one value after a transform | inspect the readback sample; an all-`0` / all-`Unknown` column means the condition or column mapping was wrong, not that the data is like that |
+| Decide on every blank the deliverable uses | one report line for each `blank_values` entry (`project check` lists them all: from every view and dashboard): the column, the count, and filled, filtered, or kept and why. Kept is a decision only when you write it down |
 | A join states its match rate and explains unmatched rows | row count before/after, `IS_EMPTY` filter count on a right-side column, and the key columns named exactly as both view schemas show them |
 | Group counts add up | Σ of per-group counts from a pivot/crosstab equals the source row count (or nulls in the group key explain the difference) |
 | Every job you submitted reached a terminal state | `job get JOB_ID` shows `success`; report anything else as failed/unknown, not done |

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/commands/dashboard.py

@@ -43,6 +43,7 @@ from mammoth_cli.services.dashboard_pages import check_added_pages
 from mammoth_cli.services.dashboard_review import (
     CANVAS_GET,
     NEW_COLUMN_NOTE,
+    columns_not_on_dashboard,
     has_profiles,
     review,
 )
@@ -721,10 +722,21 @@ def _with_deliverable_check(
             if isinstance(view_id, int)
             else None
         )
-        fallback = None
-        if not has_profiles(canvas_doc) and isinstance(view_id, int) and dataset_id is not None:
-            fallback = view_profiles(service, dataset_id, view_id, resolved_project(invocation))
-        warnings = review(canvas_doc, dataset_id, fallback)
+        current = (
+            view_profiles(service, dataset_id, view_id, resolved_project(invocation))
+            if isinstance(view_id, int) and dataset_id is not None
+            else None
+        )
+        if has_profiles(canvas_doc):
+            stale = columns_not_on_dashboard(canvas_doc, current)
+            warnings = stale + [
+                warning
+                for warning in review(canvas_doc, dataset_id)
+                # On a stale board the money advice is the rebuild, not a new column.
+                if not (stale and warning["issue"] in {"money_not_shown", "unit_price_summed"})
+            ]
+        else:
+            warnings = review(canvas_doc, dataset_id, current)
     except Exception:  # noqa: BLE001 -- advice must never fail the authoring step
         return data
     if warnings:

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/commands/doctor.py

@@ -13,6 +13,7 @@ import platform
 import re
 import shlex
 import sys
+import time
 from pathlib import Path
 from typing import Any, cast
 from urllib.parse import urlsplit
@@ -132,6 +133,25 @@ def _nearest_existing(path: Path) -> Path:
     return path
 
 
+#: ``doctor --input '{"wait": N}'``: retry a failing connection for up to N
+#: seconds (capped), probing every ``_WAIT_INTERVAL`` seconds.
+_WAIT_INTERVAL = 15
+_WAIT_MAX = 900
+
+
+def _wait_seconds(invocation: Invocation) -> float:
+    document = invocation.load_input() or {}
+    try:
+        value = float(document.get("wait") or 0)
+    except (TypeError, ValueError):
+        raise CliError(
+            code="invalid_argument",
+            message="wait must be a number of seconds.",
+            hint="mammoth doctor --input '{\"wait\": 300}'",
+        ) from None
+    return max(0.0, min(value, _WAIT_MAX))
+
+
 def doctor(invocation: Invocation) -> HandlerResult:
     """Run environment and connectivity diagnostics.
 
@@ -241,13 +261,30 @@ def doctor(invocation: Invocation) -> HandlerResult:
     connection_ok = False
     connection_detail = "not attempted"
     projects: list[dict[str, Any]] | None = None
+    wait_seconds = _wait_seconds(invocation)
     if auth_ok:
+        started = time.monotonic()
+        retries = 0
         try:
-            with open_service(invocation) as (service, _auth):
-                service.check_connection()
-                projects = _visible_projects(service)
+            while True:
+                try:
+                    with open_service(invocation) as (service, _auth):
+                        service.check_connection()
+                        projects = _visible_projects(service)
+                    break
+                except CliError as error:
+                    # A 502/504 or timeout during an outage: with ``wait``,
+                    # probe again instead of failing on the first try.
+                    remaining = wait_seconds - (time.monotonic() - started)
+                    if not error.retryable or remaining <= 0:
+                        raise
+                    time.sleep(min(_WAIT_INTERVAL, remaining))
+                    retries += 1
+            waited = time.monotonic() - started
             connection_ok = True
-            connection_detail = "authenticated request succeeded"
+            connection_detail = "authenticated request succeeded" + (
+                f" after {retries} retries over {round(waited)}s" if retries else ""
+            )
         except CliError as error:
             diagnostics = _safe_connection_diagnostics(error, debug=invocation.debug)
             connection_detail = diagnostics.pop("detail")

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/commands/project.py

@@ -509,3 +509,80 @@ def project_user_update(invocation: Invocation) -> HandlerResult:
     with open_service(invocation) as (service, auth):
         data = service.call(_symbol(invocation), **kwargs)
     return data, _meta(invocation, auth.workspace_id, project_id)
+
+
+_DATASETS_LIST_SYMBOL = "mammoth.api.datasets.DatasetsAPI.list_all"
+_DASHBOARDS_LIST_SYMBOL = "mammoth.api.dashboards.DashboardsAPI.list"
+
+
+def _report_line(where: str, warning: dict[str, Any]) -> str:
+    column = warning.get("column")
+    subject = f"{where}, column {column}" if column else where
+    return f"{subject}: {warning.get('issue')}. {warning.get('detail', '')}".strip()
+
+
+def project_check(invocation: Invocation) -> HandlerResult:
+    """List what a report on this project must still account for.
+
+    Read-only local composite, run before reporting: for each dataset, the
+    first view's ``column_warnings`` and ``before_dashboard``; for each
+    dashboard, its ``deliverable_check``. ``to_report`` flattens them into
+    one line per finding. Cold-agent evals (2.0.41) left one column's blanks
+    undecided in every run, because the warning sat in an earlier result.
+    """
+    from mammoth_cli.commands.dashboard import _with_deliverable_check
+    from mammoth_cli.commands.view import upload_preview
+
+    project_id = _project_id(invocation)
+    views: list[dict[str, Any]] = []
+    dashboards: list[dict[str, Any]] = []
+    to_report: list[str] = []
+    with open_service(invocation) as (service, auth):
+        listing = service.call(_DATASETS_LIST_SYMBOL, project_id=project_id)
+        datasets = listing.get("datasets", []) if isinstance(listing, dict) else []
+        for dataset in datasets:
+            dataset_id = dataset.get("id") if isinstance(dataset, dict) else None
+            if not isinstance(dataset_id, int):
+                continue
+            preview = upload_preview(service, dataset_id, project_id)
+            if preview is None:
+                continue
+            entry = {
+                "dataset_id": dataset_id,
+                "dataset_name": dataset.get("name"),
+                "view_id": preview["view_id"],
+                "row_count": preview.get("row_count"),
+                "column_warnings": preview.get("column_warnings", []),
+            }
+            if "before_dashboard" in preview:
+                entry["before_dashboard"] = preview["before_dashboard"]
+            views.append(entry)
+            where = f"view {preview['view_id']} ({dataset.get('name')})"
+            to_report += [_report_line(where, w) for w in entry["column_warnings"]]
+        boards = service.call(_DASHBOARDS_LIST_SYMBOL, project_id=project_id)
+        for board in boards if isinstance(boards, list) else []:
+            dashboard_id = board.get("id") if isinstance(board, dict) else None
+            if not isinstance(dashboard_id, int):
+                continue
+            checked = _with_deliverable_check(
+                invocation, service, auth, {"dashboard_id": dashboard_id}, {"id": dashboard_id}
+            )
+            check = checked.get("deliverable_check") if isinstance(checked, dict) else None
+            warnings = check.get("warnings", []) if isinstance(check, dict) else []
+            dashboards.append(
+                {"id": dashboard_id, "title": board.get("title"), "warnings": warnings}
+            )
+            to_report += [_report_line(f"dashboard {dashboard_id}", w) for w in warnings]
+        meta = _meta(invocation, auth.workspace_id, project_id)
+    return {
+        "project_id": project_id,
+        "views": views,
+        "dashboards": dashboards,
+        "to_report": to_report,
+        "note": (
+            "Before you report: fix each finding, or give it one line in the report "
+            "(what you did and why). A finding you do not mention is a miss."
+            if to_report
+            else "Nothing open in the views or dashboards."
+        ),
+    }, meta

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/commands/registry.py

@@ -205,6 +205,7 @@ HANDLERS: dict[str, Handler] = {
     # project family (read-only)
     "project.list": project_cmd.project_list,
     "project.get": project_cmd.project_get,
+    "project.check": project_cmd.project_check,
     "project.pending-changes": project_cmd.project_pending_changes,
     "project.resource-status": project_cmd.project_resource_status,
     "project.resource-dependencies": project_cmd.project_resource_dependencies,

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/services/command_contract.py

@@ -257,6 +257,7 @@ _PILOT_ADAPTER_INPUTS: dict[str, frozenset[str]] = {
 # values that may cross the structured-input boundary.
 _LOCAL_CONTRACT_FIELDS: dict[str, tuple[FieldSpec, ...]] = {
     "project.ensure": (FieldSpec("name", required=False, annotation=str),),
+    "doctor": (FieldSpec("wait", required=False, annotation=int, default=0),),
     "schema.get": (FieldSpec("full", required=False, annotation=bool, default=False),),
     "schema.list": (
         FieldSpec("family", required=False, annotation=str | None, default=None),
@@ -346,6 +347,7 @@ _LOCAL_COMMANDS = frozenset(
         "context.project.use",
         "dataset.find",
         "doctor",
+        "project.check",
         "project.ensure",
         "folder.find",
         "log.path",

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/services/dashboard_review.py

@@ -279,6 +279,43 @@ def upload_hints(
     return warnings
 
 
+def columns_not_on_dashboard(
+    canvas_doc: Any, view_columns: Iterable[Mapping[str, Any]] | None
+) -> list[dict[str, Any]]:
+    """Warn about view columns that the dashboard's profile does not have.
+
+    A dashboard sees the view as it was when the dashboard was made
+    (``NEW_COLUMN_NOTE``), so a column added later cannot be charted there.
+    ``view_columns`` is the view's own profile (``profiles_from_view``).
+    """
+    known = {p["name"] for p in _profiles(canvas_doc)} if isinstance(canvas_doc, Mapping) else set()
+    if not known or not view_columns:
+        return []
+    missing = [
+        str(c["name"])
+        for c in view_columns
+        if isinstance(c.get("name"), str) and c["name"] not in known
+    ]
+    if not missing:
+        return []
+    canvas = canvas_doc.get("canvas")
+    source = canvas.get("dataset") if isinstance(canvas, Mapping) else None
+    view_id = source.get("dataview_id") if isinstance(source, Mapping) else None
+    spec = json.dumps({"params": {"dataview_id": view_id if view_id is not None else "VIEW_ID"}})
+    return [
+        {
+            "issue": "columns_not_on_dashboard",
+            "columns": missing,
+            "detail": (
+                f"The view has {', '.join(missing)}, but this dashboard was made before "
+                f"and cannot chart {'it' if len(missing) == 1 else 'them'}. Make a new "
+                "dashboard on the view and build there."
+            ),
+            "fix": f"mammoth dashboard create-blank --input '{spec}' --yes",
+        }
+    ]
+
+
 def has_profiles(canvas_doc: Any) -> bool:
     """Whether a canvas read carries the backend's column profile."""
     return isinstance(canvas_doc, Mapping) and bool(_profiles(canvas_doc))

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/services/mapping.py

@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+import re
 import shlex
 from typing import Any
 
@@ -106,6 +107,35 @@ def _job_recovery(job_id: object, *, profile: str | None = None) -> list[str]:
     ]
 
 
+_PROJECT_DELETE = re.compile(r"/workspaces/\d+/projects/(\d+)/?$")
+_DASHBOARD_WRITE = re.compile(r"/dashboards/(\d+)(?:/|$)")
+
+
+def _resource_recovery(
+    method: object, endpoint: object, project_id: object, *, profile: str | None = None
+) -> list[str]:
+    """Reads that settle an unknown outcome for writes with no job handle.
+
+    Seen on release during an outage: ``dashboard create-blank`` and
+    ``project delete`` timed out with nothing to inspect, and agents could
+    not tell whether to replay. Each command here answers that question.
+    """
+    if not isinstance(endpoint, str):
+        return []
+    options = f"{' --profile ' + shlex.quote(profile) if profile else ''}"
+    verb = str(method or "").upper()
+    if verb == "DELETE" and (match := _PROJECT_DELETE.search(endpoint)):
+        # Deletion is asynchronous (202): read until resource_not_found.
+        return [f"mammoth project get {match.group(1)}{options}"]
+    if verb == "POST" and endpoint.rstrip("/").endswith("/dashboards/v3/blank"):
+        project = f" --project {project_id}" if isinstance(project_id, int) else ""
+        # Look for a dashboard with the title you sent before creating another.
+        return [f"mammoth dashboard list{project}{options}"]
+    if verb in {"POST", "PUT", "PATCH"} and (match := _DASHBOARD_WRITE.search(endpoint)):
+        return [f"mammoth dashboard canvas get {match.group(1)}{options}"]
+    return []
+
+
 def map_sdk_exception(
     exc: BaseException,
     *,
@@ -250,6 +280,10 @@ def map_sdk_exception(
             )
         )
         if uncertain_effect:
+            if not recovery:
+                recovery = _resource_recovery(
+                    method, details.get("endpoint"), project_id, profile=profile
+                )
             details.setdefault("operation_state", CODE_OUTCOME_UNKNOWN)
             if method is None:
                 details.setdefault("metadata_missing", ["method", "operation_state"])

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/mammoth_cli/services/positionals.py

@@ -242,6 +242,7 @@ POSITIONAL_OVERRIDES: dict[str, tuple[PositionalSpec, ...]] = {
     # active project). The SDK signature marks it required or omits it, so the
     # dual-sourced optional locator is authored here.
     "project.get": _optional_project_id(),
+    "project.check": _optional_project_id(),
     "project.pending-changes": _optional_project_id(),
     "project.resource-status": _optional_project_id(),
     "project.resource-dependencies": _optional_project_id(),

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "mammoth-cli"
-version = "2.0.41"
+version = "2.0.43"
 description = "Command-line interface for the Mammoth Analytics platform"
 authors = [{name = "Mammoth Analytics", email = "support@mammoth.io"}]
 readme = "README.md"

{mammoth_cli-2.0.41 → mammoth_cli-2.0.43}/spec/manifests/commands/project.yaml

@@ -71,6 +71,45 @@ commands:
     The CLI requires every patches[].value item to name a project_id and the confirmation names
     those ids; a body that names no project is refused before dispatch. Not verified live.
   reviewed_by: primary
+- command_id: project.check
+  command_path: project check
+  disposition: command
+  alias_of: null
+  operation_ids: []
+  positionals:
+  - name: project_id
+    type: int
+    required: false
+    metavar: PROJECT_ID
+    falls_back_to_field: null
+    help: ID of the project to act on; defaults to the active project.
+  options: []
+  sdk_symbol: mammoth_cli.commands.project.project_check
+  sdk_conversion: Call mammoth_cli.commands.project.project_check with validated request fields.
+  request_model: ProjectCheckRequest
+  result_model: ProjectCheckResult
+  mutation_class: read
+  confirmation: none
+  wait_policy: not_async
+  pagination_policy: none
+  acceptance_evidence: live_read_only
+  live_test: null
+  live_exemption_reason: Local composite over already-covered dataset, view and dashboard reads.
+  contract_fixture: null
+  required_fixture_guard: null
+  secret_fields: []
+  human_example: mammoth project check --help
+  agent_example: mammoth project check 123
+  unit_tests:
+  - UT-PROJECT-CHECK
+  contract_tests:
+  - CT-PROJECT-CHECK-HUMAN
+  - CT-PROJECT-CHECK-JSON
+  - CT-PROJECT-CHECK-ERROR
+  draft_test: null
+  undo_test: null
+  known_restrictions: Reads the first view of each dataset (the one uploads and the playbooks use) and one data page per view.
+  reviewed_by: primary
 - command_id: project.checkpoint.list
   command_path: project checkpoint list
   disposition: command