arkclaw-webchat-cli 0.9.0__tar.gz → 0.13.0__tar.gz

{arkclaw_webchat_cli-0.9.0 → arkclaw_webchat_cli-0.13.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: arkclaw-webchat-cli
-Version: 0.9.0
+Version: 0.13.0
 Summary: CLI to chat with an ArkClaw EE space's Claw over enterprise SSO — zero permanent AK/SK.
 Author: ArkClaw Team
 Keywords: arkclaw,cli,ee,openclaw,sso,sts
@@ -49,6 +49,7 @@ arkclaw chat ci-xxxxxxxx
 - **Upload & download any file** — push arbitrary files into the claw's *real* workspace filesystem and pull results back (a whole directory comes as a `.tar`) with `upload` / `download` / `mkdir` / `ls <path>`, over the Silk file service (needs a Silk-enabled claw).
 - **Large-product storage** — a separate IDS *netdisk* plane for big artifacts/videos: `netdisk create-space` / `ls` / `upload` / `download` / `mkdir` / `rm`.
 - **Read past conversations** — list an agent's sessions, then pull a full transcript with `arkclaw history <会话ID>` (over the `chat.history` ws RPC; `--json` to export, `-o` to a file).
+- **Schedule tasks** — `arkclaw cron` makes the claw run a prompt unattended on a cron expression / fixed interval / one-shot time (`add` / `list` / `run` / `runs` / `update` / `rm`), over the runtime's `cron.*` RPCs.
 - **Fan out** — send one message to many claws in parallel.
 - **Be scripted** — every command takes `--json` (one clean `{ok,data,error}` envelope on stdout) and returns a typed exit code.
 
@@ -118,14 +119,14 @@ straight to `arkclaw login https://your-space...`. See **[Configuration](#config
 | `arkclaw upload <local> [remote]` | Upload **any** file into the claw's workspace filesystem (Silk). Defaults to the basename at the workspace root. |
 | `arkclaw download <remote> [local]` | Download a workspace file from the claw (Silk); a directory comes as `<name>.tar`. |
 | `arkclaw mkdir <path>` | Create a directory in the claw workspace (Silk; parent must exist). |
-| `arkclaw netdisk <cmd>` | IDS netdisk (large products/videos), a separate storage plane: `create-space` / `ls` / `upload` / `download` / `mkdir` / `rm`. Configured via `IDS_*` env. |
+| `arkclaw netdisk <cmd>` | IDS netdisk (large products/videos), a separate storage plane: `create-space` / `ls` / `upload` / `download` / `mkdir` / `rm`. Per-user creds minted from your SSO login (no static AK/SK). |
 | `arkclaw fanout "<msg>" --clawid ci-a --clawid ci-b` | Same message to many claws in parallel. |
 | `arkclaw sessions [claw]` | An agent's conversations from the server (`--agent` to pick one), newest first — resume with `chat --session <会话ID>`, read with `history <会话ID>`, fork with `--new`. Falls back to the local record when offline / a2a. |
-| `arkclaw history <session>` | Print a past conversation's full transcript (ws `chat.history`). `<session>` = a 会话ID from `sessions`, a label, or a full `agent:…` key. Tool steps hidden unless `--show-tools`; `--json` dumps every message; `-o` writes a plain-text log. openclaw only. |
+| `arkclaw history <session>` | Print a past conversation's transcript (ws `chat.history`). `<session>` = a 会话ID from `sessions`, a label, or a full `agent:…` key. The screen shows the last `--tail N` messages (default 50; `--all` for everything); tool steps hidden unless `--show-tools`; `--json` dumps every message; `-o` writes a plain-text log. openclaw only. |
+| `arkclaw cron <cmd>` | Scheduled tasks the claw runs unattended (ws `cron.*`): `status` (is the scheduler on), `list`, `get <id>`, `add --name … (--cron EXPR \| --every DUR \| --at TIME) -m "<prompt>"`, `update <id>`, `run <id>` (fire now), `runs [id]` (history), `rm <id>`. Needs a claw whose runtime has the scheduler enabled (`cron status`). openclaw only. |
 | `arkclaw profile save/use/list` | Named snapshots of the session config (multi-space / multi-claw). |
 | `arkclaw doctor` | Self-check: login freshness, keychain, pool/STS/endpoint reachability. |
 | `arkclaw schema --json` | Machine-readable command surface (for agents/tooling). |
-| `arkclaw claw list/describe/create/delete` | Fleet management on the control plane. |
 
 All commands accept `--clawid ci-...` (where relevant) to target a specific claw, and `--json` for machine output.
 
@@ -155,7 +156,7 @@ The CLI moves files across **three distinct planes** — pick by what you're car
 |---|---|---|
 | **Managed brain** | `ls` · `pull` · `push` | The claw's own config files (`AGENTS.md`, `SOUL.md`, …). A fixed allow-list, not a general store. |
 | **Workspace (Silk)** | `ls <path>` · `upload` · `download` · `mkdir` | The claw's *real* working filesystem — arbitrary files, uploads, agent outputs, subdirectories. |
-| **Netdisk (IDS)** | `netdisk …` | A separate large-object store for big products/videos, on a dedicated storage account. |
+| **Netdisk (IDS)** | `netdisk …` | A separate large-object store for big products/videos; per-user creds minted from your SSO login. |
 
 ### 1. Managed brain files — `ls` / `pull` / `push`
 
@@ -195,26 +196,58 @@ arkclaw download runs ./runs.tar       # a directory downloads as a .tar
 
 ### 3. Netdisk (IDS) — `netdisk …`
 
-A separate storage plane for large products/videos, on a **dedicated IDS
-account** (not your SSO identity). Configured entirely from `IDS_*` env vars
-(provisioned by the storage team). Create a space once, then pass its `SpaceID`:
+A separate storage plane for large products/videos. Credentials are **minted
+per-user from your SSO login** — no static AK/SK: the claw's iDrive instance is
+resolved and a minimal-permission temp credential is issued for *your*
+`UserPoolUserUid` (`GetIDriveInstance` → `GetIDriveCredential`). `--space`
+defaults to the claw's iDrive space; pass it to target another.
 
 ```bash
-arkclaw netdisk create-space my-space            # → prints the SpaceID
-arkclaw netdisk mkdir outputs --space spc-...
-arkclaw netdisk upload ./final.mp4 outputs/final.mp4 --space spc-...
-arkclaw netdisk ls outputs/ --space spc-...
-arkclaw netdisk download outputs/final.mp4 ./final.mp4 --space spc-...
-arkclaw netdisk rm outputs/final.mp4 --space spc-...   # asks unless --yes
+arkclaw netdisk ls                                # the claw's iDrive space
+arkclaw netdisk upload ./final.mp4 outputs/final.mp4
+arkclaw netdisk download outputs/final.mp4 ./final.mp4
+arkclaw netdisk mkdir outputs
+arkclaw netdisk rm outputs/final.mp4              # asks unless --yes
+arkclaw netdisk create-space my-space             # a second space → prints its SpaceID
+arkclaw netdisk ls --space spc-...                # target a specific space
 ```
 
 `create-space` also takes `--description`; `ls` takes `--limit N` to cap entries.
 
+For CI / a dedicated storage account you can still override the identity path:
+
 | Env | Purpose |
 |---|---|
-| `IDS_BASE_URL` / `IDS_REGION` / `IDS_SERVICE` / `IDS_INSTANCE_ID` | Endpoint + routing for the IDS instance. |
-| `IDS_AK` / `IDS_SK` (+ `IDS_ACCOUNT_ID`) | Static credentials for the storage account… |
-| `IDS_ROLE_TRN` | …or assume a role via SSO instead of static AK/SK. |
+| `IDS_ROLE_TRN` (+ `IDS_BASE_URL` / `IDS_REGION` / `IDS_SERVICE` / `IDS_INSTANCE_ID`) | Assume an IDS role via SSO instead. |
+| `IDS_AK` / `IDS_SK` (+ `IDS_ACCOUNT_ID`, endpoint/routing vars) | Static credentials for a dedicated storage account. |
+
+---
+
+## Scheduled tasks — `cron`
+
+Let a claw run a prompt **unattended** on a schedule (its runtime carries a
+scheduler; check it with `arkclaw cron status`). A task is an *agent turn* — the
+`--message` is the prompt the agent runs at each fire.
+
+```bash
+arkclaw cron status                         # is the scheduler on; jobs; next wake
+arkclaw cron add --name 早报 \
+  --cron "0 9 * * *" --tz Asia/Shanghai \
+  -m "联网搜集今天的科技新闻,生成一份中文摘要"   # every day 09:00
+arkclaw cron add --name 巡检 --every 30m -m "检查服务健康并报告异常"
+arkclaw cron add --name 一次性 --at 2026-07-01T09:00:00+08:00 -m "提醒我项目截止" --delete-after-run
+arkclaw cron list                           # all jobs (enabled + disabled)
+arkclaw cron run <id>                        # fire one now (server-side)
+arkclaw cron runs <id>                       # recent run history + status
+arkclaw cron update <id> --disable           # pause; --enable to resume
+arkclaw cron rm <id>                         # delete (asks unless --yes)
+```
+
+Schedule is exactly one of `--cron "<expr>"` (with optional `--tz`), `--every
+<30s|15m|2h|1d>`, or `--at <ISO8601>`. `--session-target` picks where the run
+lives (`isolated` default / `main` / `current`); `--agent` runs it as a specific
+agent. Jobs the platform manages (created by an agent-pack template) are marked
+`*` in the list — edit those in the console, not here. openclaw plane only.
 
 ---
 
@@ -238,8 +271,10 @@ arkclaw netdisk rm outputs/final.mp4 --space spc-...   # asks unless --yes
 ## Configuration
 
 `login` resolves config in this order: **explicit flag → `ARKCLAW_*` env →
-the space's published discovery → `arkclaw init` defaults → derived from the
-address**. You normally only type the address.
+the space's discovery → `arkclaw init` defaults → derived from the address**.
+You normally only type the address — when the space exposes the CLI bootstrap
+endpoint, `login` auto-fetches its public OAuth client + STS role, so even
+`--client-id` / `--role-trn` are optional.
 
 **The easy way — `arkclaw init`** asks once (interactively) for whatever the
 space doesn't yet auto-publish, and saves it. After that, just `arkclaw login`.
@@ -253,6 +288,7 @@ values via env:
 | `ARKCLAW_ROLE_TRN` | STS role that mints the ChatToken (openclaw). |
 | `ARKCLAW_REGION` / `ARKCLAW_SPACE_ID` | Override region / space (usually auto). |
 | `ARKCLAW_API_HOST` / `ARKCLAW_API_SERVICE` | **Advanced/testing.** Point the control-plane calls at a non-production plane (host + SigV4 service). See the note below. |
+| `ARKCLAW_FORWARD_ENV` | **Advanced/non-prod.** Adds an `X-Forward-Env` header to the space's CLI-bootstrap call to route to a small-traffic gateway. Unset (production) sends no header. |
 
 Other config: `~/.arkclaw/defaults.json` (`init` answers, 0600), `~/.arkclaw/session.json` (routing, 0600), `~/.arkclaw/cmdpolicy.json` (tool-approval allow/deny), OS keychain (tokens).
 

{arkclaw_webchat_cli-0.9.0 → arkclaw_webchat_cli-0.13.0}/README.md

@@ -24,6 +24,7 @@ arkclaw chat ci-xxxxxxxx
 - **Upload & download any file** — push arbitrary files into the claw's *real* workspace filesystem and pull results back (a whole directory comes as a `.tar`) with `upload` / `download` / `mkdir` / `ls <path>`, over the Silk file service (needs a Silk-enabled claw).
 - **Large-product storage** — a separate IDS *netdisk* plane for big artifacts/videos: `netdisk create-space` / `ls` / `upload` / `download` / `mkdir` / `rm`.
 - **Read past conversations** — list an agent's sessions, then pull a full transcript with `arkclaw history <会话ID>` (over the `chat.history` ws RPC; `--json` to export, `-o` to a file).
+- **Schedule tasks** — `arkclaw cron` makes the claw run a prompt unattended on a cron expression / fixed interval / one-shot time (`add` / `list` / `run` / `runs` / `update` / `rm`), over the runtime's `cron.*` RPCs.
 - **Fan out** — send one message to many claws in parallel.
 - **Be scripted** — every command takes `--json` (one clean `{ok,data,error}` envelope on stdout) and returns a typed exit code.
 
@@ -93,14 +94,14 @@ straight to `arkclaw login https://your-space...`. See **[Configuration](#config
 | `arkclaw upload <local> [remote]` | Upload **any** file into the claw's workspace filesystem (Silk). Defaults to the basename at the workspace root. |
 | `arkclaw download <remote> [local]` | Download a workspace file from the claw (Silk); a directory comes as `<name>.tar`. |
 | `arkclaw mkdir <path>` | Create a directory in the claw workspace (Silk; parent must exist). |
-| `arkclaw netdisk <cmd>` | IDS netdisk (large products/videos), a separate storage plane: `create-space` / `ls` / `upload` / `download` / `mkdir` / `rm`. Configured via `IDS_*` env. |
+| `arkclaw netdisk <cmd>` | IDS netdisk (large products/videos), a separate storage plane: `create-space` / `ls` / `upload` / `download` / `mkdir` / `rm`. Per-user creds minted from your SSO login (no static AK/SK). |
 | `arkclaw fanout "<msg>" --clawid ci-a --clawid ci-b` | Same message to many claws in parallel. |
 | `arkclaw sessions [claw]` | An agent's conversations from the server (`--agent` to pick one), newest first — resume with `chat --session <会话ID>`, read with `history <会话ID>`, fork with `--new`. Falls back to the local record when offline / a2a. |
-| `arkclaw history <session>` | Print a past conversation's full transcript (ws `chat.history`). `<session>` = a 会话ID from `sessions`, a label, or a full `agent:…` key. Tool steps hidden unless `--show-tools`; `--json` dumps every message; `-o` writes a plain-text log. openclaw only. |
+| `arkclaw history <session>` | Print a past conversation's transcript (ws `chat.history`). `<session>` = a 会话ID from `sessions`, a label, or a full `agent:…` key. The screen shows the last `--tail N` messages (default 50; `--all` for everything); tool steps hidden unless `--show-tools`; `--json` dumps every message; `-o` writes a plain-text log. openclaw only. |
+| `arkclaw cron <cmd>` | Scheduled tasks the claw runs unattended (ws `cron.*`): `status` (is the scheduler on), `list`, `get <id>`, `add --name … (--cron EXPR \| --every DUR \| --at TIME) -m "<prompt>"`, `update <id>`, `run <id>` (fire now), `runs [id]` (history), `rm <id>`. Needs a claw whose runtime has the scheduler enabled (`cron status`). openclaw only. |
 | `arkclaw profile save/use/list` | Named snapshots of the session config (multi-space / multi-claw). |
 | `arkclaw doctor` | Self-check: login freshness, keychain, pool/STS/endpoint reachability. |
 | `arkclaw schema --json` | Machine-readable command surface (for agents/tooling). |
-| `arkclaw claw list/describe/create/delete` | Fleet management on the control plane. |
 
 All commands accept `--clawid ci-...` (where relevant) to target a specific claw, and `--json` for machine output.
 
@@ -130,7 +131,7 @@ The CLI moves files across **three distinct planes** — pick by what you're car
 |---|---|---|
 | **Managed brain** | `ls` · `pull` · `push` | The claw's own config files (`AGENTS.md`, `SOUL.md`, …). A fixed allow-list, not a general store. |
 | **Workspace (Silk)** | `ls <path>` · `upload` · `download` · `mkdir` | The claw's *real* working filesystem — arbitrary files, uploads, agent outputs, subdirectories. |
-| **Netdisk (IDS)** | `netdisk …` | A separate large-object store for big products/videos, on a dedicated storage account. |
+| **Netdisk (IDS)** | `netdisk …` | A separate large-object store for big products/videos; per-user creds minted from your SSO login. |
 
 ### 1. Managed brain files — `ls` / `pull` / `push`
 
@@ -170,26 +171,58 @@ arkclaw download runs ./runs.tar       # a directory downloads as a .tar
 
 ### 3. Netdisk (IDS) — `netdisk …`
 
-A separate storage plane for large products/videos, on a **dedicated IDS
-account** (not your SSO identity). Configured entirely from `IDS_*` env vars
-(provisioned by the storage team). Create a space once, then pass its `SpaceID`:
+A separate storage plane for large products/videos. Credentials are **minted
+per-user from your SSO login** — no static AK/SK: the claw's iDrive instance is
+resolved and a minimal-permission temp credential is issued for *your*
+`UserPoolUserUid` (`GetIDriveInstance` → `GetIDriveCredential`). `--space`
+defaults to the claw's iDrive space; pass it to target another.
 
 ```bash
-arkclaw netdisk create-space my-space            # → prints the SpaceID
-arkclaw netdisk mkdir outputs --space spc-...
-arkclaw netdisk upload ./final.mp4 outputs/final.mp4 --space spc-...
-arkclaw netdisk ls outputs/ --space spc-...
-arkclaw netdisk download outputs/final.mp4 ./final.mp4 --space spc-...
-arkclaw netdisk rm outputs/final.mp4 --space spc-...   # asks unless --yes
+arkclaw netdisk ls                                # the claw's iDrive space
+arkclaw netdisk upload ./final.mp4 outputs/final.mp4
+arkclaw netdisk download outputs/final.mp4 ./final.mp4
+arkclaw netdisk mkdir outputs
+arkclaw netdisk rm outputs/final.mp4              # asks unless --yes
+arkclaw netdisk create-space my-space             # a second space → prints its SpaceID
+arkclaw netdisk ls --space spc-...                # target a specific space
 ```
 
 `create-space` also takes `--description`; `ls` takes `--limit N` to cap entries.
 
+For CI / a dedicated storage account you can still override the identity path:
+
 | Env | Purpose |
 |---|---|
-| `IDS_BASE_URL` / `IDS_REGION` / `IDS_SERVICE` / `IDS_INSTANCE_ID` | Endpoint + routing for the IDS instance. |
-| `IDS_AK` / `IDS_SK` (+ `IDS_ACCOUNT_ID`) | Static credentials for the storage account… |
-| `IDS_ROLE_TRN` | …or assume a role via SSO instead of static AK/SK. |
+| `IDS_ROLE_TRN` (+ `IDS_BASE_URL` / `IDS_REGION` / `IDS_SERVICE` / `IDS_INSTANCE_ID`) | Assume an IDS role via SSO instead. |
+| `IDS_AK` / `IDS_SK` (+ `IDS_ACCOUNT_ID`, endpoint/routing vars) | Static credentials for a dedicated storage account. |
+
+---
+
+## Scheduled tasks — `cron`
+
+Let a claw run a prompt **unattended** on a schedule (its runtime carries a
+scheduler; check it with `arkclaw cron status`). A task is an *agent turn* — the
+`--message` is the prompt the agent runs at each fire.
+
+```bash
+arkclaw cron status                         # is the scheduler on; jobs; next wake
+arkclaw cron add --name 早报 \
+  --cron "0 9 * * *" --tz Asia/Shanghai \
+  -m "联网搜集今天的科技新闻,生成一份中文摘要"   # every day 09:00
+arkclaw cron add --name 巡检 --every 30m -m "检查服务健康并报告异常"
+arkclaw cron add --name 一次性 --at 2026-07-01T09:00:00+08:00 -m "提醒我项目截止" --delete-after-run
+arkclaw cron list                           # all jobs (enabled + disabled)
+arkclaw cron run <id>                        # fire one now (server-side)
+arkclaw cron runs <id>                       # recent run history + status
+arkclaw cron update <id> --disable           # pause; --enable to resume
+arkclaw cron rm <id>                         # delete (asks unless --yes)
+```
+
+Schedule is exactly one of `--cron "<expr>"` (with optional `--tz`), `--every
+<30s|15m|2h|1d>`, or `--at <ISO8601>`. `--session-target` picks where the run
+lives (`isolated` default / `main` / `current`); `--agent` runs it as a specific
+agent. Jobs the platform manages (created by an agent-pack template) are marked
+`*` in the list — edit those in the console, not here. openclaw plane only.
 
 ---
 
@@ -213,8 +246,10 @@ arkclaw netdisk rm outputs/final.mp4 --space spc-...   # asks unless --yes
 ## Configuration
 
 `login` resolves config in this order: **explicit flag → `ARKCLAW_*` env →
-the space's published discovery → `arkclaw init` defaults → derived from the
-address**. You normally only type the address.
+the space's discovery → `arkclaw init` defaults → derived from the address**.
+You normally only type the address — when the space exposes the CLI bootstrap
+endpoint, `login` auto-fetches its public OAuth client + STS role, so even
+`--client-id` / `--role-trn` are optional.
 
 **The easy way — `arkclaw init`** asks once (interactively) for whatever the
 space doesn't yet auto-publish, and saves it. After that, just `arkclaw login`.
@@ -228,6 +263,7 @@ values via env:
 | `ARKCLAW_ROLE_TRN` | STS role that mints the ChatToken (openclaw). |
 | `ARKCLAW_REGION` / `ARKCLAW_SPACE_ID` | Override region / space (usually auto). |
 | `ARKCLAW_API_HOST` / `ARKCLAW_API_SERVICE` | **Advanced/testing.** Point the control-plane calls at a non-production plane (host + SigV4 service). See the note below. |
+| `ARKCLAW_FORWARD_ENV` | **Advanced/non-prod.** Adds an `X-Forward-Env` header to the space's CLI-bootstrap call to route to a small-traffic gateway. Unset (production) sends no header. |
 
 Other config: `~/.arkclaw/defaults.json` (`init` answers, 0600), `~/.arkclaw/session.json` (routing, 0600), `~/.arkclaw/cmdpolicy.json` (tool-approval allow/deny), OS keychain (tokens).
 

{arkclaw_webchat_cli-0.9.0 → arkclaw_webchat_cli-0.13.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "arkclaw-webchat-cli"
-version = "0.9.0"
+version = "0.13.0"
 description = "CLI to chat with an ArkClaw EE space's Claw over enterprise SSO — zero permanent AK/SK."
 readme = "README.md"
 requires-python = ">=3.10"

{arkclaw_webchat_cli-0.9.0 → arkclaw_webchat_cli-0.13.0}/src/ee_claw/cli.py

@@ -238,16 +238,27 @@ def history(
     show_tools: bool = typer.Option(
         False, "--show-tools", help="Also show tool calls/results (default: conversation only)."
     ),
+    tail: int = typer.Option(
+        flows.DEFAULT_HISTORY_TAIL,
+        "--tail",
+        "-n",
+        help=f"Show only the last N messages on screen (default {flows.DEFAULT_HISTORY_TAIL}).",
+    ),
+    all_messages: bool = typer.Option(
+        False, "--all", help="Show the whole transcript on screen (no tail cap)."
+    ),
     json_mode: bool = _json_opt(),
 ) -> None:
-    """Read a past conversation's full transcript (ws `chat.history`). Get the
-    会话ID from `arkclaw sessions`, then `arkclaw history <会话ID>`. `--json`
-    returns every message (incl. tool steps); `-o` writes a plain-text log."""
+    """Read a past conversation's transcript (ws `chat.history`). Get the 会话ID
+    from `arkclaw sessions`, then `arkclaw history <会话ID>`. The screen shows the
+    last `--tail` messages (default 50; `--all` for everything); `-o` and `--json`
+    always carry the full transcript (incl. tool steps for `--json`)."""
     emitter = Emitter(json_mode=json_mode)
     _run(
         emitter,
         lambda: flows.do_history(
-            emitter, session, clawid=clawid, agent=agent, output=output, show_tools=show_tools
+            emitter, session, clawid=clawid, agent=agent, output=output,
+            show_tools=show_tools, tail=tail, show_all=all_messages,
         ),
     )
 
@@ -417,14 +428,200 @@ def fanout(
     )
 
 
-def _space_opt() -> str:
-    return typer.Option(..., "--space", help="Netdisk SpaceID (from `netdisk create-space`).")
+cron_app = typer.Typer(
+    no_args_is_help=True,
+    help="定时任务:让 claw 按计划(cron 表达式 / 固定间隔 / 指定时间)自动跑一段 agent 任务 "
+    "(无人值守)。需要运行时开了调度器的 claw —— `arkclaw cron status` 先确认。",
+)
+app.add_typer(cron_app, name="cron")
+
+
+def _cron_clawid() -> str | None:
+    return typer.Option(
+        None, "--clawid", help="Claw instance id (ci-...). Omit to use your default claw."
+    )
+
+
+@cron_app.command("status")
+def cron_status(clawid: str | None = _cron_clawid(), json_mode: bool = _json_opt()) -> None:
+    """Is this claw's scheduler running, how many jobs, and when it next wakes."""
+    emitter = Emitter(json_mode=json_mode)
+    _run(emitter, lambda: flows.do_cron_status(emitter, clawid=clawid))
+
+
+@cron_app.command("list")
+def cron_list(clawid: str | None = _cron_clawid(), json_mode: bool = _json_opt()) -> None:
+    """List the claw's scheduled jobs (enabled and disabled). `cron get <id>`
+    for one job's full definition; `cron run <id>` to fire it now."""
+    emitter = Emitter(json_mode=json_mode)
+    _run(emitter, lambda: flows.do_cron_list(emitter, clawid=clawid))
+
+
+@cron_app.command("get")
+def cron_get(
+    job_id: str = typer.Argument(..., metavar="ID", help="Job id from `cron list`."),
+    clawid: str | None = _cron_clawid(),
+    json_mode: bool = _json_opt(),
+) -> None:
+    """Show one scheduled job's full definition + last-run state."""
+    emitter = Emitter(json_mode=json_mode)
+    _run(emitter, lambda: flows.do_cron_get(emitter, job_id, clawid=clawid))
+
+
+@cron_app.command("add")
+def cron_add(
+    name: str = typer.Option(..., "--name", help="任务名(显示用)。"),
+    message: str = typer.Option(
+        ..., "--message", "-m", help="到点让 agent 执行的提示词(无人值守的一段任务)。"
+    ),
+    cron_expr: str | None = typer.Option(
+        None, "--cron", help='Cron 表达式,如 "0 9 * * *"(每天9点)。与 --every/--at 三选一。'
+    ),
+    every: str | None = typer.Option(
+        None, "--every", help="固定间隔:30s / 15m / 2h / 1d。与 --cron/--at 三选一。"
+    ),
+    at: str | None = typer.Option(
+        None, "--at", help="只跑一次的时间点(ISO8601),如 2026-07-01T09:00:00+08:00。"
+    ),
+    tz: str | None = typer.Option(
+        None, "--tz", help='时区(配合 --cron),如 Asia/Shanghai。'
+    ),
+    agent: str | None = typer.Option(
+        None, "--agent", help="用哪个 agent 执行(agentId;默认 claw 的主 agent)。"
+    ),
+    model: str | None = typer.Option(None, "--model", help="覆盖执行用的模型(可选)。"),
+    timeout: int | None = typer.Option(
+        None, "--timeout", help="单次运行的超时(秒,可选)。"
+    ),
+    session_target: str = typer.Option(
+        "isolated",
+        "--session-target",
+        help="会话目标:isolated(每次独立,默认)/ main(主会话)/ current。",
+    ),
+    disabled: bool = typer.Option(
+        False, "--disabled", help="创建后先不启用(默认启用)。"
+    ),
+    delete_after_run: bool = typer.Option(
+        False, "--delete-after-run", help="跑完一次就删除(常配合 --at)。"
+    ),
+    clawid: str | None = _cron_clawid(),
+    json_mode: bool = _json_opt(),
+) -> None:
+    """Create a scheduled agent-turn task. Give a schedule (exactly one of
+    --cron / --every / --at) and the --message the agent runs unattended.
+    Example: `arkclaw cron add --name 早报 --cron "0 9 * * *" -m "生成今日新闻摘要"`."""
+    emitter = Emitter(json_mode=json_mode)
+    _run(
+        emitter,
+        lambda: flows.do_cron_add(
+            emitter,
+            name=name,
+            message=message,
+            cron_expr=cron_expr,
+            every=every,
+            at=at,
+            tz=tz,
+            agent=agent,
+            model=model,
+            timeout=timeout,
+            session_target=session_target,
+            enabled=not disabled,
+            delete_after_run=delete_after_run,
+            clawid=clawid,
+        ),
+    )
+
+
+@cron_app.command("update")
+def cron_update(
+    job_id: str = typer.Argument(..., metavar="ID", help="Job id from `cron list`."),
+    name: str | None = typer.Option(None, "--name", help="新任务名。"),
+    message: str | None = typer.Option(None, "--message", "-m", help="新的任务提示词。"),
+    cron_expr: str | None = typer.Option(None, "--cron", help='新 cron 表达式。'),
+    every: str | None = typer.Option(None, "--every", help="新间隔:30s/15m/2h/1d。"),
+    at: str | None = typer.Option(None, "--at", help="新的一次性时间点(ISO8601)。"),
+    tz: str | None = typer.Option(None, "--tz", help="时区(配合 --cron)。"),
+    enable: bool = typer.Option(False, "--enable", help="启用该任务。"),
+    disable: bool = typer.Option(False, "--disable", help="停用该任务。"),
+    clawid: str | None = _cron_clawid(),
+    json_mode: bool = _json_opt(),
+) -> None:
+    """Patch a job — only the fields you pass change (name / schedule / message /
+    enabled)."""
+    emitter = Emitter(json_mode=json_mode)
+    if enable and disable:
+        raise typer.Exit(
+            emitter.fail(ValidationError("--enable 和 --disable 不能同时给。"))
+        )
+    enable_val: bool | None = True if enable else (False if disable else None)
+    _run(
+        emitter,
+        lambda: flows.do_cron_update(
+            emitter,
+            job_id,
+            name=name,
+            message=message,
+            cron_expr=cron_expr,
+            every=every,
+            at=at,
+            tz=tz,
+            enable=enable_val,
+            clawid=clawid,
+        ),
+    )
+
+
+@cron_app.command("rm")
+def cron_rm(
+    job_id: str = typer.Argument(..., metavar="ID", help="Job id from `cron list`."),
+    yes: bool = typer.Option(False, "--yes", help="Skip the confirmation prompt."),
+    clawid: str | None = _cron_clawid(),
+    json_mode: bool = _json_opt(),
+) -> None:
+    """Delete a scheduled job (asks for confirmation unless --yes)."""
+    emitter = Emitter(json_mode=json_mode)
+    _run(emitter, lambda: flows.do_cron_remove(emitter, job_id, clawid=clawid, yes=yes))
+
+
+@cron_app.command("run")
+def cron_run(
+    job_id: str = typer.Argument(..., metavar="ID", help="Job id from `cron list`."),
+    due: bool = typer.Option(
+        False, "--due", help="只在当前确实到点时才跑(默认强制立即跑)。"
+    ),
+    clawid: str | None = _cron_clawid(),
+    json_mode: bool = _json_opt(),
+) -> None:
+    """Fire a scheduled job now (runs server-side; `cron runs <id>` for results)."""
+    emitter = Emitter(json_mode=json_mode)
+    _run(emitter, lambda: flows.do_cron_run(emitter, job_id, clawid=clawid, due_only=due))
+
+
+@cron_app.command("runs")
+def cron_runs(
+    job_id: str | None = typer.Argument(
+        None, metavar="[ID]", help="Job id; omit for the whole claw's history."
+    ),
+    limit: int = typer.Option(10, "--limit", "-n", help="Max entries to show."),
+    clawid: str | None = _cron_clawid(),
+    json_mode: bool = _json_opt(),
+) -> None:
+    """Recent run history — one job (with ID) or the whole claw (without)."""
+    emitter = Emitter(json_mode=json_mode)
+    _run(emitter, lambda: flows.do_cron_runs(emitter, job_id, clawid=clawid, limit=limit))
+
+
+def _space_opt() -> str | None:
+    return typer.Option(
+        None, "--space",
+        help="Netdisk SpaceID. Defaults to the claw's iDrive space; set to target another.",
+    )
 
 
 netdisk_app = typer.Typer(
     no_args_is_help=True,
-    help="IDS 网盘 (large products / videos): a separate storage plane. Configure "
-    "via IDS_* env vars (base_url/account/ak/sk/region/service/instance, from the storage team).",
+    help="IDS 网盘 (large products / videos): a separate storage plane. Credentials are "
+    "minted per-user from your SSO login (no static AK/SK). IDS_* env still override for CI.",
 )
 app.add_typer(netdisk_app, name="netdisk")
 
@@ -443,7 +640,7 @@ def netdisk_create_space(
 @netdisk_app.command("ls")
 def netdisk_ls(
     prefix: str = typer.Argument("", help="Dir prefix to list (e.g. 'outputs/'). Omit for root."),
-    space: str = _space_opt(),
+    space: str | None = _space_opt(),
     limit: int | None = typer.Option(None, "--limit", help="Max entries to return."),
     json_mode: bool = _json_opt(),
 ) -> None:
@@ -456,7 +653,7 @@ def netdisk_ls(
 def netdisk_upload(
     local: str = typer.Argument(..., help="Local file to upload."),
     remote: str | None = typer.Argument(None, help="Netdisk path (default: basename)."),
-    space: str = _space_opt(),
+    space: str | None = _space_opt(),
     json_mode: bool = _json_opt(),
 ) -> None:
     """Upload a file into a netdisk space."""
@@ -468,7 +665,7 @@ def netdisk_upload(
 def netdisk_download(
     remote: str = typer.Argument(..., help="Netdisk file path to download."),
     local: str | None = typer.Argument(None, help="Local path (default: same name here)."),
-    space: str = _space_opt(),
+    space: str | None = _space_opt(),
     json_mode: bool = _json_opt(),
 ) -> None:
     """Download a netdisk file (streamed to disk)."""
@@ -479,7 +676,7 @@ def netdisk_download(
 @netdisk_app.command("mkdir")
 def netdisk_mkdir(
     path: str = typer.Argument(..., help="Netdisk folder path to create."),
-    space: str = _space_opt(),
+    space: str | None = _space_opt(),
     json_mode: bool = _json_opt(),
 ) -> None:
     """Create a folder in a netdisk space."""
@@ -490,7 +687,7 @@ def netdisk_mkdir(
 @netdisk_app.command("rm")
 def netdisk_rm(
     remote: str = typer.Argument(..., help="Netdisk file path to delete."),
-    space: str = _space_opt(),
+    space: str | None = _space_opt(),
     yes: bool = typer.Option(False, "--yes", help="Skip the confirmation prompt."),
     json_mode: bool = _json_opt(),
 ) -> None:
@@ -499,53 +696,6 @@ def netdisk_rm(
     _run(emitter, lambda: flows.do_netdisk_rm(emitter, space, remote, yes=yes))
 
 
-claw_app = typer.Typer(
-    no_args_is_help=True,
-    help="Fleet: manage claw instances on the control plane.",
-)
-app.add_typer(claw_app, name="claw")
-
-
-@claw_app.command("list")
-def claw_list(json_mode: bool = _json_opt()) -> None:
-    """List claw instances in the logged-in space's project."""
-    emitter = Emitter(json_mode=json_mode)
-    _run(emitter, lambda: flows.do_claw(emitter, "list"))
-
-
-@claw_app.command("describe")
-def claw_describe(
-    clawid: str = typer.Argument(..., help="Claw instance id (ci-...)."),
-    json_mode: bool = _json_opt(),
-) -> None:
-    """Describe one claw instance."""
-    emitter = Emitter(json_mode=json_mode)
-    _run(emitter, lambda: flows.do_claw(emitter, "describe", clawid=clawid))
-
-
-@claw_app.command("create")
-def claw_create(
-    body: str = typer.Option(
-        ..., "--body", help='Raw create request JSON, e.g. \'{"Name": "my-claw"}\'.'
-    ),
-    json_mode: bool = _json_opt(),
-) -> None:
-    """Create a claw instance (raw control-plane body passthrough)."""
-    emitter = Emitter(json_mode=json_mode)
-    _run(emitter, lambda: flows.do_claw(emitter, "create", body=body))
-
-
-@claw_app.command("delete")
-def claw_delete(
-    clawid: str = typer.Argument(..., help="Claw instance id (ci-...)."),
-    yes: bool = typer.Option(False, "--yes", help="Skip the confirmation prompt."),
-    json_mode: bool = _json_opt(),
-) -> None:
-    """Delete a claw instance (asks for confirmation unless --yes)."""
-    emitter = Emitter(json_mode=json_mode)
-    _run(emitter, lambda: flows.do_claw(emitter, "delete", clawid=clawid, yes=yes))
-
-
 @app.command()
 def doctor(json_mode: bool = _json_opt()) -> None:
     """Self-check: config, login freshness, keychain, pool/STS/endpoint