用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/tomevault-io/skills-registry --skill doover-admin命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
| name | doover-admin |
| description | Administration guide for the Doover platform Use when this capability is needed. |
| metadata | {"author":"getdoover"} |
This skill covers administration topics for the Doover platform, including organization setup, device types, and deployment configuration.
Organizations are the top-level entity in Doover, containing users, devices, and applications.
Organization
├── Users (members with roles)
├── Device Types (templates for devices)
├── Devices (physical/virtual devices)
├── Applications (apps deployed to devices)
└── API Keys (for programmatic access)
| Role | Permissions |
|---|---|
| Owner | Full access, billing, can delete org |
| Admin | Manage users, devices, apps |
| Developer | Deploy apps, view devices |
| Viewer | Read-only access to dashboards |
Device types define templates for devices with pre-configured applications.
Device types specify:
Key fields for device type integration:
{
"my_app": {
"key": "uuid-goes-here",
"name": "my_app",
"display_name": "My Application",
"type": "DEV",
"visibility": "PUB",
"allow_many": true,
"description": "Short description",
"long_description": "README.md",
"depends_on": ["platform_interface"],
"owner_org_key": "org-uuid",
"image_name": "ghcr.io/getdoover/my_app",
"container_registry_profile_key": "",
"build_args": "--platform linux/amd64,linux/arm64"
}
}
| Field | Type | Description |
|---|---|---|
key | UUID | Unique identifier for the app |
name | string | Internal name (snake_case) |
display_name | string | Human-readable name |
type | enum | DEV (development) or PROD (production) |
visibility | enum | PUB (public) or PRI (private) |
allow_many | boolean | Allow multiple instances on one device |
description | string | Short description |
long_description | string | Path to README or long description |
depends_on | array | Required system apps |
owner_org_key | UUID | Owning organization |
image_name | string | Docker image registry path |
container_registry_profile_key | UUID | Registry credentials profile |
build_args | string | Docker build arguments |
Applications can depend on system services:
{
"depends_on": [
"platform_interface",
"device_agent",
"modbus_interface"
]
}
| Dependency | Purpose |
|---|---|
platform_interface | GPIO, hardware I/O access |
device_agent | Channel publishing, tag management |
modbus_interface | Modbus RTU/TCP communication |
Provides access to device hardware:
class MyApplication(Application):
async def main_loop(self):
# Digital input
values = await self.platform_iface.get_di_async([1, 2, 3])
# Digital output
await self.platform_iface.set_do_async(pin=4, value=True)
# Analog input (if supported)
analog = await self.platform_iface.get_ai_async([1])
Provides data management services:
class MyApplication(Application):
async def main_loop(self):
# Publish to channel
await self.device_agent.publish_to_channel_async(
"sensor_data",
json.dumps({"temperature": 25.5})
)
For Modbus communication:
class MyApplication(Application):
async def main_loop(self):
# Read holding registers
registers = await self.modbus_iface.read_registers_async(
start_address=100,
count=10,
register_type=3 # Holding registers
)
# Write register
await self.modbus_iface.write_register_async(
address=200,
value=1
)
Configure container registry access in doover_config.json:
{
"my_app": {
"image_name": "ghcr.io/getdoover/my_app",
"container_registry_profile_key": "registry-profile-uuid"
}
}
Supported registries:
Multi-platform builds for ARM and x86 devices:
{
"my_app": {
"build_args": "--platform linux/amd64,linux/arm64"
}
}
Common build arguments:
--platform linux/amd64 - Intel/AMD 64-bit--platform linux/arm64 - ARM 64-bit (Raspberry Pi 4, etc.)--platform linux/arm/v7 - ARM 32-bit (older Pi models)# 1. Ensure you're authenticated
doover auth login
# 2. Build for target platforms
doover app build
# 3. Publish to platform
doover app publish
# For staging/testing
doover app publish --staging
# Production with specific profile
doover app publish --profile production
Applications receive configuration via environment variables:
| Variable | Description |
|---|---|
APP_KEY | Unique key for this app instance |
CONFIG_FP | Path to configuration JSON file |
HEALTHCHECK_PORT | Port for health check endpoint |
import os
class MyApplication(Application):
async def setup(self):
app_key = os.environ.get("APP_KEY")
config_path = os.environ.get("CONFIG_FP")
log.info(f"Starting app {app_key}")
log.info(f"Config from {config_path}")
For local development:
services:
my_app:
build: ../
network_mode: host
environment:
- APP_KEY=test_app_key
- CONFIG_FP=/app/simulators/app_config.json
| Visibility | Description |
|---|---|
PUB | Public - visible to all organizations |
PRI | Private - only visible to owner organization |
In doover_config.json:
{
"my_app": {
"visibility": "PRI",
"owner_org_key": "your-org-uuid"
}
}
Private apps can be shared with specific organizations through the Doover platform.
Applications include health check endpoints:
HEALTHCHECK --interval=30s --timeout=2s --start-period=5s \
CMD curl -f "127.0.0.1:$HEALTHCHECK_PORT" || exit 1
Monitor application status via tags:
class MyApplication(Application):
async def main_loop(self):
# Report health status
await self.set_tag("health", {
"status": "healthy",
"uptime_seconds": self.uptime,
"last_error": None,
"version": "1.0.0"
})
class MyApplication(Application):
async def main_loop(self):
try:
await self.do_work()
await self.set_tag("status", "running")
await self.set_tag("last_error", None)
except Exception as e:
log.error(f"Work failed: {e}")
await self.set_tag("status", "error")
await self.set_tag("last_error", str(e))
await self.set_tag("last_error_time", datetime.now().isoformat())
Use profiles for different environments:
# Default profile
doover app publish
# Staging profile
doover app publish --profile staging
# Production profile
doover app publish --profile production
Configure profiles in Doover CLI:
# Add a profile
doover config add-profile staging --api-url https://staging.doover.com
# List profiles
doover config list-profiles
# Switch default profile
doover config set-profile production
Use semantic versioning for apps:
{
"metadata": {
"version": "1.2.3"
}
}
__version__ = "1.2.3"
class MyApplication(Application):
async def setup(self):
await self.set_tag("app_version", __version__)
Access logs from deployed devices:
# Via Docker on device
ssh device "docker logs my_app --tail 100"
# Follow logs
ssh device "docker logs my_app -f"
Enable debug mode via configuration:
class MyApplication(Application):
async def main_loop(self):
if self.config.debug_mode.value:
await self.debug_log("main_loop", {
"temperature": self.temp,
"state": self.state.state
})
async def debug_log(self, context: str, data: dict):
await self.device_agent.publish_to_channel_async(
"debug",
json.dumps({
"timestamp": datetime.now().isoformat(),
"context": context,
"data": data
})
)
Use the channel viewer for real-time debugging:
doover app channels --host device-ip
pyproject.tomldoover app rundoover app testdoover app publishKeep previous versions available:
For high-traffic applications:
Converted and distributed by TomeVault — claim your Tome and manage your conversions.
基于 SOC 职业分类