dot-device-openapi
Interact with Dot. devices through OpenAPI - list devices, check status, switch content, and send Text/Image/Canvas API payloads.
How do I install this agent skill?
npx skills add https://github.com/mindreset/dot_skill --skill dot-device-openapiIs this agent skill safe to install?
- Gen Agent Trust Hubpass
The skill provides tools and scripts to interact with Dot. devices via the MindReset OpenAPI. It supports listing devices, checking status, updating settings, and sending content like text, images, or canvas payloads. It uses standard API key authentication through environment variables and provides instructions for installing a local MCP server from the vendor's official GitHub repository. No security issues were detected.
- Socketwarn
1 alert: gptAnomaly
- Snykwarn
Risk: MEDIUM · 1 issue
What does this agent skill do?
Dot Device OpenAPI and MCP
Use this skill when the user wants to operate a Dot. device:
- List devices or find a device ID
- Check device status
- Read or update device settings
- Switch to the next content
- Send Text API content
- Send Image API content
- Send an already-designed Canvas API payload
- List loop or fixed tasks on a device
If the user asks to design a custom Canvas card, dashboard, or windowData layout, use dot-canvas-designer first, then return here only when the user wants to send it to a device.
Product Routing
Read products.md before choosing an integration for Read Pico or Rand/0, or when the user asks for hardware resources. Read Pico development uses its official firmware repository; Rand/0 Display Mode uses a local image protocol. Do not request Dot cloud credentials or use the sending scripts for those routes. The requirements below apply to supported Dot cloud API devices.
Local MCP
Compatible agents can use the local dot-mcp stdio server instead of constructing HTTP requests. It exposes only the fixed Dot tools below; it does not expose arbitrary HTTP access:
- Read-only:
dot_list_devices,dot_get_device_status,dot_get_device_settings,dot_list_timezones,dot_list_tasks,dot_validate_canvas - Device-changing:
dot_update_device_settings,dot_switch_next_content,dot_send_text,dot_send_image,dot_send_canvas
The server reads DOT_API_KEY from the MCP process environment. It never accepts the key as a tool argument, stores it, or logs it. Device-changing requests are sent once without automatic retries.
Install the local server with pipx install git+https://github.com/MindReset/dot_skill.git, run it with dot-mcp, or use uvx --from git+https://github.com/MindReset/dot_skill.git dot-mcp. Platform configuration examples are in docs/mcp-configs.md.
Requirements
- Base URL:
https://dot.mindreset.tech - Authentication:
Authorization: Bearer dot_app_<api_key> - API key source: ask the user or read
DOT_API_KEYfrom the local environment. - Device ID source: ask the user, list devices, or use the ID the user provides.
- Rate limit: 10 requests per second.
Successful POST control endpoints (next, text, image, canvas) return a JSON object with a top-level message field. Do not expect a legacy { code, message, result } wrapper.
Workflow
- Confirm a valid API key and device ID.
- For Text API, Image API, or Canvas API writes, ensure the matching content has already been added to the device loop task in Dot. App Content Studio.
- Pick the endpoint:
- Text display:
POST /api/authV2/open/device/:deviceId/text - Image display:
POST /api/authV2/open/device/:deviceId/image - Canvas display:
POST /api/authV2/open/device/:deviceId/canvas - Device status:
GET /api/authV2/open/device/:deviceId/status - Device settings:
GET /api/authV2/open/device/:deviceId/settings - Update device settings:
POST /api/authV2/open/device/:deviceId/settings - Timezone list:
GET /api/authV2/open/timezones - Device list:
GET /api/authV2/open/devices - Next content:
POST /api/authV2/open/device/:deviceId/next - Task list:
GET /api/authV2/open/device/:deviceId/:taskType/list
- Text display:
- Always include
Content-Type: application/jsonfor POST requests.
Scripts
Use the scripts in scripts/ for local execution:
list_devices.pyget_device_status.pyget_device_settings.pyupdate_device_settings.pylist_tasks.pyswitch_next.pysend_text.pysend_image.pysend_canvas.py
Text, image, and Canvas helper scripts support --task-alias for the user-readable task name shown in the device task list.
Request Notes
For Text API, parameters include refreshNow, taskKey, taskAlias, title, message, signature, icon, link, and styles.
For Image API, parameters include refreshNow, taskKey, taskAlias, image, link, border, ditherType, and ditherKernel.
For Canvas API sending, parameters include refreshNow, taskKey, taskAlias, data, windowData, layoutFull, link, and border. Build or revise windowData with dot-canvas-designer.
For device settings, parameters include alias, location, timezone, interval, and sleep. Timezones must be one of the keys returned by GET /api/authV2/open/timezones. Both interval.powerMs and interval.batteryMs must be 60,000-43,200,000 ms in whole-minute multiples. The battery interval controls automatic wake and content refresh timing. sleep.start and sleep.end use local HH:mm time in the device timezone, and an end time earlier than start means the next day.
For Text API, Image API, and Canvas API, use top-level taskAlias when the user wants a human-readable task-list name. Omit taskAlias to keep the existing task name. Send taskAlias: "" or taskAlias: null only when the user explicitly wants to clear the name. Never put taskAlias inside Canvas data.
References
references/authentication.mdreferences/api_reference.md- OpenAPI schema:
../../openapi/dot-openapi.yaml
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/mindreset/dot_skill/dot-device-openapi">View dot-device-openapi on skillZs</a>