mammoth-cli 2.0.35__tar.gz → 2.0.37__tar.gz

{mammoth_cli-2.0.35 → mammoth_cli-2.0.37}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mammoth-cli
-Version: 2.0.35
+Version: 2.0.37
 Summary: Command-line interface for the Mammoth Analytics platform
 Project-URL: homepage, https://mammoth.io
 Project-URL: repository, https://github.com/EdgeMetric/mammothsdk
@@ -17,7 +17,7 @@ Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Requires-Python: <3.15,>=3.12
 Requires-Dist: jsonschema>=4.20
 Requires-Dist: keyring<26,>=25.7
-Requires-Dist: mammoth-io<0.8,>=0.7.15
+Requires-Dist: mammoth-io<0.8,>=0.7.16
 Requires-Dist: platformdirs<5,>=4.10
 Requires-Dist: pydantic<3,>=2.13
 Requires-Dist: pyyaml<7,>=6
@@ -203,21 +203,24 @@ one-time login, and works from the shipped skill. The long form is
 [docs/agent-prompt.md](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/agent-prompt.md).
 
 ```text
-Use Mammoth Analytics only through the `mammoth` CLI in bash; run
-`export MAMMOTH_OUTPUT=json MAMMOTH_NO_INPUT=1` once. Onboard me first:
+Use Mammoth Analytics only through the `mammoth` CLI in bash, never the web
+app in a browser; run `export MAMMOTH_OUTPUT=json MAMMOTH_NO_INPUT=1` once.
+Onboard me first:
 1. If `mammoth --version` fails, install, then re-check (new shell if needed):
    curl -fsSL https://raw.githubusercontent.com/EdgeMetric/mammothsdk/main/mammoth-cli/installers/mammoth-install.sh | bash
-2. Run `mammoth auth status`. If it shows no credentials, print exactly this
-   and wait until I say done: "In the Mammoth web app open Workspace settings
-   → API Tokens → Create token, copy the key and the secret (shown once), and
-   note the workspace id (the number after /workspaces/ in the address bar),
-   then run in your own terminal: mammoth auth login" — never ask for, read, or pass a key
-   or secret yourself, and never run auth login.
+2. Run `mammoth auth status`. If it shows no credentials or a
+   `recommendation`, print exactly this and wait until I say done: "In the
+   Mammoth web app open Workspace settings → API Tokens → Create token (no
+   project, so I can create projects), copy the token (it starts with mm_ and
+   is shown once), and note the workspace id (the number after /workspaces/
+   in the address bar), then run in your own terminal: mammoth auth login" —
+   never ask for, read, or pass the token yourself, and never run auth login.
 3. Require `mammoth doctor` to pass, then run `mammoth skill show` and
    follow the guide it prints.
 Work inside `mammoth project ensure 'PROJECT NAME'` unless I name a project;
-take ids only from reads; `schema get COMMAND_ID` before a new command; read
-results back before reporting.
+take ids only from reads; find a command with `schema find "GOAL IN PLAIN
+WORDS"` or `view transform --help`, and `schema get COMMAND_ID` before a new
+command; read results back before reporting.
 TASK: <what to achieve, and how you will know it is done>
 ```
 
@@ -265,7 +268,7 @@ The full generated list is in [docs/reference/commands.md](https://github.com/Ed
 |---|---|
 | [Installation](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/installation.md) | Install the CLI and the agent skill. |
 | [Quick start](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/quickstart.md) | Log in and run your first commands. |
-| [Authentication](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/authentication.md) | Getting an API key, login, profiles, projects. |
+| [Authentication](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/authentication.md) | Getting an API token, login, profiles, projects. |
 | [Agent handover and operation](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/agents.md) | Cold start, discovery, checkpoints, recovery. |
 | [Portable handoff format](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/agent-handoff.md) | Nonsecret checkpoint schema and receiving procedure. |
 | [Bundled agent skill](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/mammoth_cli/bundled_skill/mammoth-cli/SKILL.md) | Focused routing for shell-capable agents. |

{mammoth_cli-2.0.35 → mammoth_cli-2.0.37}/README.md

@@ -175,21 +175,24 @@ one-time login, and works from the shipped skill. The long form is
 [docs/agent-prompt.md](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/agent-prompt.md).
 
 ```text
-Use Mammoth Analytics only through the `mammoth` CLI in bash; run
-`export MAMMOTH_OUTPUT=json MAMMOTH_NO_INPUT=1` once. Onboard me first:
+Use Mammoth Analytics only through the `mammoth` CLI in bash, never the web
+app in a browser; run `export MAMMOTH_OUTPUT=json MAMMOTH_NO_INPUT=1` once.
+Onboard me first:
 1. If `mammoth --version` fails, install, then re-check (new shell if needed):
    curl -fsSL https://raw.githubusercontent.com/EdgeMetric/mammothsdk/main/mammoth-cli/installers/mammoth-install.sh | bash
-2. Run `mammoth auth status`. If it shows no credentials, print exactly this
-   and wait until I say done: "In the Mammoth web app open Workspace settings
-   → API Tokens → Create token, copy the key and the secret (shown once), and
-   note the workspace id (the number after /workspaces/ in the address bar),
-   then run in your own terminal: mammoth auth login" — never ask for, read, or pass a key
-   or secret yourself, and never run auth login.
+2. Run `mammoth auth status`. If it shows no credentials or a
+   `recommendation`, print exactly this and wait until I say done: "In the
+   Mammoth web app open Workspace settings → API Tokens → Create token (no
+   project, so I can create projects), copy the token (it starts with mm_ and
+   is shown once), and note the workspace id (the number after /workspaces/
+   in the address bar), then run in your own terminal: mammoth auth login" —
+   never ask for, read, or pass the token yourself, and never run auth login.
 3. Require `mammoth doctor` to pass, then run `mammoth skill show` and
    follow the guide it prints.
 Work inside `mammoth project ensure 'PROJECT NAME'` unless I name a project;
-take ids only from reads; `schema get COMMAND_ID` before a new command; read
-results back before reporting.
+take ids only from reads; find a command with `schema find "GOAL IN PLAIN
+WORDS"` or `view transform --help`, and `schema get COMMAND_ID` before a new
+command; read results back before reporting.
 TASK: <what to achieve, and how you will know it is done>
 ```
 
@@ -237,7 +240,7 @@ The full generated list is in [docs/reference/commands.md](https://github.com/Ed
 |---|---|
 | [Installation](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/installation.md) | Install the CLI and the agent skill. |
 | [Quick start](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/quickstart.md) | Log in and run your first commands. |
-| [Authentication](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/authentication.md) | Getting an API key, login, profiles, projects. |
+| [Authentication](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/authentication.md) | Getting an API token, login, profiles, projects. |
 | [Agent handover and operation](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/agents.md) | Cold start, discovery, checkpoints, recovery. |
 | [Portable handoff format](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/docs/agent-handoff.md) | Nonsecret checkpoint schema and receiving procedure. |
 | [Bundled agent skill](https://github.com/EdgeMetric/mammothsdk/blob/main/mammoth-cli/mammoth_cli/bundled_skill/mammoth-cli/SKILL.md) | Focused routing for shell-capable agents. |

{mammoth_cli-2.0.35 → mammoth_cli-2.0.37}/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__ = "2.0.35"
+__version__ = "2.0.37"
 
 # The versioned machine output envelope contract.
 SCHEMA_VERSION = 1

{mammoth_cli-2.0.35 → mammoth_cli-2.0.37}/mammoth_cli/bundled_skill/mammoth-cli/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: mammoth-cli
-version: 2.0.35
-description: "Use Mammoth Analytics from a terminal: install or authenticate the CLI, discover its live command contract, and safely manage projects, data, views, pipelines, dashboards, exports, and handoffs."
+version: 2.0.37
+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."
 ---
 
 # Mammoth CLI
@@ -10,6 +10,19 @@ Use this skill for Mammoth shell work, not for Python SDK integration. The
 CLI is the contract: it validates every request locally, returns one JSON
 envelope, and needs no flag for that when you pipe its output.
 
+## What Mammoth is
+
+Mammoth Analytics is a no-code data platform. Data comes in from files, URLs
+or connectors and lands in a **dataset**. It is cleaned, joined, reshaped and
+calculated in a **pipeline** of tasks on a **view**, and the result is
+published as a dashboard or delivered as a CSV, a database table, a BI feed
+or another dataset. The pipeline re-runs when new data arrives, so build the
+steps in Mammoth rather than computing a result locally. The web app and
+this CLI call the same API: every task in the web app's Transform menu is a
+`mammoth view transform` command. The data model, the high-level
+capabilities and the full web-to-CLI map are in
+[about Mammoth](references/about-mammoth.md).
+
 ## Start
 
 ```bash
@@ -23,7 +36,7 @@ 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
 `--profile NAME` only for a profile other than `default`; [auth](references/auth.md)
-says where the key and secret come from) and wait; never ask for a key or secret in chat,
+says where the token comes from) and wait; never ask for the token in chat,
 never read one from a file or environment variable, never run `auth login`
 yourself. If any envelope carries `meta.update_available`, run its `command`
 before the next step.
@@ -42,6 +55,38 @@ before the next step.
 - Commands that start platform work wait up to 300 s for the job; on
   `timeout` use the `job get` recovery command printed, do not resubmit.
 
+## Find the command for a goal
+
+State the goal in plain words; the search knows common phrasing, and
+`view transform --help` is the Transform menu with one line per task:
+
+```bash
+mammoth schema find "merge two datasets"   # -> view.transform.join
+mammoth view transform --help              # every data transformation
+```
+
+| Goal | Command |
+|---|---|
+| Combine two datasets on a key (merge, VLOOKUP) | `view transform join`; one value per key: `view transform lookup` |
+| Add rows to an existing dataset | `file upload FILE --input '{"append_to_ds_id": DATASET_ID}'` |
+| Keep or remove rows | `view transform filter` |
+| Remove duplicate rows | `view transform discard-duplicates` |
+| Totals, counts, averages per group | `view transform pivot` |
+| A calculated column | `view transform math` |
+| Blanks to a constant, or to the previous row's value | `view transform set-values` (`IS_EMPTY`), `view transform fill-missing` |
+| Clean text | `view transform text`, `replace`, `bulk-replace` |
+| Change a column's type | `view transform convert-type` |
+| Rank, running total, previous row | `view transform window` |
+| A dashboard | `dashboard create-blank` ([dashboards](references/recipes/dashboards.md)) |
+| Deliver the rows | `view export csv`, `view export postgres` (and other destinations), `view export dataset` |
+
+A search with no full match returns `suggestions` and a `hint`. Before you
+conclude the CLI cannot do something the web app does, check
+`view transform --help` and "Not a pipeline step" in
+[about Mammoth](references/about-mammoth.md). Do not switch to the web app
+in a browser, or compute the result locally, for work the CLI covers;
+report a real gap with the command you tried.
+
 ## Calling a command
 
 - Ids are positionals; request fields are one `--input '{...}'` document;
@@ -89,6 +134,8 @@ before the next step.
 
 ## Route only what the task needs
 
+- **What Mammoth is, its data model, which feature fits a business goal, or
+  a web-app task you cannot find in the CLI:** [about Mammoth](references/about-mammoth.md)
 - **Cold start, task plan, or unsupported route:** [task start](references/task-start.md)
 - **Login, profiles, and authorized scope:** [auth](references/auth.md)
 - **Nested input, output envelopes, confirmations, jobs, or recovery:**

mammoth_cli-2.0.37/mammoth_cli/bundled_skill/mammoth-cli/references/about-mammoth.md

@@ -0,0 +1,144 @@
+# What Mammoth is, and what it can do
+
+Read this when the task is stated as a business goal ("combine these files",
+"a monthly summary per region", "feed this to Power BI") and you need to know
+which Mammoth feature does it, or when you think the CLI cannot do something
+the web app can.
+
+## Why Mammoth exists
+
+Mammoth Analytics is a cloud platform that turns raw business data into
+output people can trust and that stays current, without code. It has two
+layers:
+
+- **Preparation (the engine).** Data comes in from files, URLs or
+  connectors. It is cleaned, reshaped, joined and calculated in a
+  **pipeline**: an ordered list of tasks on a **view**. The pipeline re-runs
+  when new data arrives, so a monthly report is built once and refreshed,
+  not rebuilt.
+- **Output.** Dashboards (and presentations, documents and Q&A pages) are
+  built from a prepared view. Data is also delivered: CSV downloads,
+  databases, cloud storage, BI tools, or another Mammoth dataset. Every
+  number is computed from the prepared data, never written by a model.
+
+People use it to replace manual spreadsheet work: an analyst automating a
+recurring report, a small practice producing client deliverables, a team
+feeding clean data to Tableau or Power BI. Every step is visible and
+repeatable, which is the point: someone else can inspect, re-run or take
+over the work. So, as an agent, build the steps **in Mammoth** (as pipeline
+tasks, dashboards and exports) rather than computing a result locally and
+uploading it; a local result does not refresh and cannot be inspected.
+
+## How Mammoth organises data
+
+```
+workspace            account and billing; an API token belongs to one workspace
+└─ project           a container with its own members; work stays inside one
+   └─ dataset        data from an upload, URL or connector; each import is a batch
+      └─ view        a working copy of the dataset's rows; a dataset can have several
+         ├─ pipeline tasks run top to bottom; each `view transform` adds one at the end
+         ├─ exports   CSV, databases, cloud storage, BI tools, another dataset
+         └─ dashboards built from the view's prepared rows
+```
+
+- An upload returns a **dataset** id. Transforms, joins, exports and
+  dashboards take a **view** id (`view list DATASET_ID`).
+- Columns have a **display name** (what people see, and what every CLI
+  input uses) and an internal name (`column_7`), which you never type.
+- A task that ran is part of the view: later data refreshes run it again.
+  Draft mode (`view draft enter` / `submit`) holds several tasks and runs
+  them together.
+
+## What it can do, and where that is in the CLI
+
+| Goal | Web app | CLI |
+|---|---|---|
+| Bring data in | Data → new dataset | `file upload`, `dataset create` (URL), `connector ...` |
+| Add rows to existing data | a new batch on the dataset | `file upload FILE --input '{"append_to_ds_id": DATASET_ID}'` |
+| Clean and reshape | View → Transform menu | `view transform OPERATION` (table below) |
+| Combine two sources | Transform → Join, Lookup | `view transform join`, `view transform lookup` |
+| Summarise | Transform → Group & Pivot | `view transform pivot`, `view transform crosstab` |
+| Check data quality in the pipeline | Data Check, Checkpoint Alert | `view data-check create`, `view checkpoint create` |
+| Reuse a value or a SQL block | Parameters, Snippets | `parameter ...`, `snippet ...` |
+| Inspect or restore pipeline history | Version history | `view version list`, `view version apply` |
+| Build a dashboard | Publish | `dashboard create-blank`, `dashboard v3 generate`, `dashboard canvas ...` |
+| Deliver data | Export, Send to dataset | `view export csv`, `view export postgres` (and other destinations), `view export dataset` |
+| Automate | Automations | `automation ...`, `workflow ...` |
+| Undo a delete | Trash | `trash list`, `trash restore` |
+
+Whether a route ran on release is in [capabilities](capabilities.md); read
+it before you promise a deliverable. `schedule` is not implemented on
+release.
+
+## The Transform menu, command by command
+
+The web app groups its pipeline tasks in a Transform menu. Each data task is
+a `mammoth view transform` command; `mammoth view transform --help` lists
+them with one line each, and `mammoth schema get view.transform.join` (or
+any other operation) gives the input. Send to Dataset is `view export dataset`; Data Check and
+Checkpoint Alert are the `view data-check` and `view checkpoint` families.
+
+| You want to | Web app task | Command |
+|---|---|---|
+| Keep or remove rows that match a condition | Conditional Filter | `filter` (`filter_type` `SHOW` or `REMOVE`) |
+| Remove duplicate rows | Remove Duplicate Rows | `discard-duplicates` |
+| Keep the first or last N rows by an order | Show Top/Bottom Rows | `limit-rows` (`n`, `order_by`, `bottom`) |
+| Write a value where a condition holds, or a label or bucket column | Label & Insert Values | `set-values` |
+| Fill blanks with the previous or next row's value | Fill Missing Values | `fill-missing` |
+| Find and replace one value | Find & Replace | `replace` |
+| Map many spellings to one value | Bulk Replace | `bulk-replace` |
+| Change case or trim spaces | Text Formatting | `text` |
+| Take part of a text value (characters or regex) | Extract Text | `substring` |
+| Split one column into several | Split Into Multiple Columns | `split` |
+| Join several columns into one text column | Combine Columns & Text | `combine-columns` |
+| Change a column's type | Convert Column Type | `convert-type` |
+| Calculate a column from others | Math Function | `math` |
+| Rank, running total, previous or next row | Window function | `window` |
+| Nth smallest or largest across columns | Obtain Large or Small Values | `small-large` |
+| Take a year, month, weekday from a date | Extract Date Part | `extract-date` |
+| Days (or months, ...) between two dates | Calculate Date Difference | `date-diff` |
+| Shift a date by an interval | Add / Subtract Date Values | `increment-date` |
+| Add an empty column | Add Column | `add-column` |
+| Copy a column under a new name | Copy Columns | `copy-columns` |
+| Remove columns | Remove Column | `delete-columns` |
+| Add columns from another view by matching keys | Join | `join` (`INNER`, `LEFT`, `RIGHT`, `OUTER`) |
+| Bring one value per key from another view | Lookup | `lookup` |
+| Totals, counts, averages per group | Group & Pivot | `pivot` (replaces the view's columns) |
+| A rows-by-columns summary table | Group & Pivot (crosstab) | `crosstab` (writes a new dataset) |
+| Turn wide columns into label/value rows | Unpivot | `unnest` |
+| Read fields out of a JSON column | Extract JSON | `json-extract` |
+| Classify, tag or summarise rows with AI | Generative AI | `ai` |
+| One SQL query over the view | AI SQL Query | `add-sql` (replaces the view's columns), `generate-sql` (from plain words) |
+| Copy rows into another dataset, refreshed with the pipeline | Send to Dataset | `view export dataset` (not a `view transform`) |
+
+`join` or `lookup`: `join` adds the chosen columns from every matching row
+(a key that repeats on the other side repeats rows); `lookup` brings one
+value per key into one new column. Before either, confirm the key columns
+in both views have the same type and values; recipes/transforms.md has the
+read-back.
+
+## Not a pipeline step, and the nearest route
+
+- **Rename a column.** The web app's rename is a display change, not a
+  task; the CLI has no rename command. Name new columns when you create them
+  (`new_column`, `as_name`); for an existing column, `copy-columns` under
+  the new name.
+- **Sort.** There is no sort task, and the CLI has no sort command. For
+  "top N by X" use `limit-rows` with `order_by`; for a rank column use
+  `window` (`RANK`, `ROW_NUMBER`).
+- **Union or append two views.** There is no union task. Upload a file into
+  an existing dataset with `append_to_ds_id`, or send a view's rows into an
+  existing dataset with `view export dataset` (`target_ds_id` and
+  `save_as_mode` `APPEND_TO_DS`; that mode is untried on release).
+- **Hide or reorder columns.** Display changes in the web app; they do not
+  change the data.
+
+## The web app and the CLI
+
+Both call the same API. Do not switch to the web app in a browser for a
+task the CLI covers. Before you conclude that the CLI cannot do something,
+run `mammoth view transform --help`, search in plain words (`mammoth schema
+find "merge two datasets"` finds `join`; a search with no full match lists
+`suggestions`), and read the section above. If it is still missing, report
+the gap with the command you tried; do not work around it in the browser or
+by computing the result locally.

{mammoth_cli-2.0.35 → mammoth_cli-2.0.37}/mammoth_cli/bundled_skill/mammoth-cli/references/auth.md

@@ -1,7 +1,16 @@
 # Authentication and profiles
 
-Required: API key, API secret, workspace id. Optional: a one-label server
-prefix (default `app`, resolving to `https://app.mammoth.io/api/v2`).
+Required: an API token (`mm_...`) and a workspace id. Optional: a one-label
+server prefix (default `app`, resolving to `https://app.mammoth.io/api/v2`).
+Login accepts nothing else. If `auth status` reports `credential: key_secret`
+(a profile saved by an older CLI), it still works, but ask the operator to run
+`mammoth auth login` again with a token.
+
+A token may be limited to one project. It then gets HTTP 403
+`authorization_required` in every other project, including a project it has
+created. When that happens, stop creating projects: ask the operator which
+project the token belongs to and pass `--project ID`, or ask for a token
+created without a project.
 
 ## Precedence
 1. Explicit login handed to the current command (secure prompt or `--input`).
@@ -13,14 +22,14 @@ history, or unrelated files for them.
 
 ## How an agent obtains credentials
 
-An agent never receives the key or secret. When `auth status` reports a
+An agent never receives the token. When `auth status` reports a
 missing profile or `has_credentials=false`, stop and hand the login to the
 human operator:
 
 1. Tell the operator where the credentials come from: in the Mammoth web app,
-   **Workspace settings → API Tokens → Create token** gives a key and a secret;
-   the web app shows the secret only once (on an existing token, **Generate new
-   secret** replaces it). The workspace id is the number after `/workspaces/`
+   **Workspace settings → API Tokens → Create token** gives a token that
+   starts with `mm_`; the web app shows it only once (if it is lost, create a
+   new one). The workspace id is the number after `/workspaces/`
    in the web app's address bar. Then give the exact command to run in
    **their own** terminal. For production it is:
 
@@ -30,9 +39,9 @@ human operator:
 
    Add `--server-prefix release` only when the task names the release
    environment, and `--profile NAME` only when the operator wants a profile
-   other than `default`. The command prompts for the key and secret with
-   hidden input, then the workspace id, so no secret enters chat, argv, or
-   history. Credentials are per environment: a release key never
+   other than `default`. The command prompts for the token with hidden
+   input, then the workspace id, so no secret enters chat, argv, or history.
+   Tokens are per environment and per workspace: a release token never
    authenticates on `app`.
    On macOS the Keychain may show a dialog about `mammoth-cli`; tell the
    operator to choose **Always Allow**. If the keychain cannot be used (for
@@ -41,12 +50,11 @@ human operator:
 2. Wait for the operator to confirm, then re-run
    `mammoth auth status --profile PROFILE` and
    `mammoth doctor --profile PROFILE`.
-3. Do not ask the operator to paste the key or secret into chat. Do not run
+3. Do not ask the operator to paste the token into chat. Do not run
    `auth login` yourself, with or without `--input`, unless the operator has
    explicitly given you a protected `0600` credential file path to use.
 
-The only supported configuration is the API key, API secret, workspace id, and
-an optional one-label server prefix (default `app`). There is no base-url
+The only supported configuration is the API token, the workspace id, and an optional one-label server prefix (default `app`). There is no base-url
 override.
 
 ## Cold-start sequence
@@ -107,8 +115,7 @@ authorization failure. When using the selected profile, `--profile PROFILE`
 may be omitted.
 
 For an ordinary human-operated terminal, prefer the first command: it asks
-for the key and secret without putting either in shell history, argv, chat, or
-logs. For headless POSIX work, create the JSON file outside the repository, set
+for the token without putting it in shell history, argv, chat, or logs. For headless POSIX work, create the JSON file outside the repository, set
 its permissions to `0600`, and pass only its path. With a protected POSIX file,
 specify `--storage file` explicitly; `--storage auto` can require an available
 OS keyring and is not a reliable headless fallback. On Windows, do not use the
@@ -116,8 +123,8 @@ file backend as a secret store; use the OS keyring or stop and obtain a
 supported secure store. Remove any input file after the
 profile is stored.
 
-`creds.json` is a `0600` JSON file: `{"api_key": "...", "api_secret": "...",
-"workspace_id": 4, "server_prefix": "app"}` (`server_prefix` optional). You can
+`creds.json` is a `0600` JSON file: `{"api_token": "mm_...", "workspace_id": 4,
+"server_prefix": "app"}` (`server_prefix` optional). You can
 also pipe it with `--input - --input-format json`.
 
 Secrets live in the OS keyring (or an explicitly selected `0600` file

{mammoth_cli-2.0.35 → mammoth_cli-2.0.37}/mammoth_cli/bundled_skill/mammoth-cli/references/handoff.md

@@ -24,7 +24,7 @@ workspace/project scope. Then add:
 - completed actions and remaining objectives;
 - a cleanup owner plus dependency order.
 
-Never include API keys, API secrets, tokens, headers, credentials files, raw
+Never include API tokens, headers, credentials files, raw
 secret-bearing input, or unredacted response bodies. IDs and display names are
 safe only when needed to identify the authorized task.
 

{mammoth_cli-2.0.35 → mammoth_cli-2.0.37}/mammoth_cli/bundled_skill/mammoth-cli/references/task-start.md

@@ -46,14 +46,17 @@ or secrets in an argument, prompt, transcript, or checkpoint.
 
    Do not print, echo, or copy secret-bearing inputs. A failed status, login,
    or doctor check stops the task before discovery and business commands.
-4. Discover instead of guessing. Run `mammoth schema list/find/get --output
-   json --no-input` for local CLI routes; `mammoth capability list --output
-   json --no-input` is an API-binding inventory and can omit typed/local
-   routes. Check [capabilities](capabilities.md) for whether a route is
-   proven, not supported, or untried on release before building a plan on
-   it. Then resolve workspace/project/dataset/view
+4. Discover instead of guessing. Search with the goal in plain words
+   (`mammoth schema find "merge two datasets"` finds the join), then run
+   `mammoth schema get COMMAND_ID` for local CLI routes; `mammoth view transform
+   --help` lists every data transformation; `mammoth schema list` is the
+   complete inventory. `mammoth capability list` is an API-binding inventory
+   and can omit typed/local routes. Check [capabilities](capabilities.md)
+   for whether a route is proven, not supported, or untried on release
+   before building a plan on it. Then resolve workspace/project/dataset/view
    parents with reads. Use display names returned by the exact view schema.
-5. Translate the business intent into a plan the agent chooses. Submit only
+5. Translate the business intent into a plan the agent chooses;
+   [about Mammoth](about-mammoth.md) maps goals to features. Submit only
    supported operations against the active project (`project ensure` sets it;
    `--project` overrides) with structured `--input`. Do not use private
    HTTP/SDK escape hatches or local data processing as a substitute.