JJR IoT Data Query and Reporting

Problem

Many IoT query workflows require manually assembling client_id, client_secret, productKey, and deviceName, while also handling token expiry, pagination, property data, subdevices, and stale image URLs. The JJR IoT Skill targets the JJR IoT platform and wraps these APIs into agent-callable capabilities: querying device lists, retrieving historical temperature/humidity/CO2/light data, generating chart reports, and pushing results to DingTalk.

How it works

• Authentication and caching: Uses https://gateway.jjr.vip/iot/ as the API base and obtains a Bearer token via POST /iot/oauth2/token; tokens last about 30 minutes, so local caching is recommended.
• Device and product queries: Supports device lists, device details, product lists, subproduct/subdevice lists, and property data queries by productKey and deviceName.
• Config and service operations: Can fetch device config, edit config, push config in batch, and invoke device services.
• Chart reports: generate_report_generic.py generates HTML and PNG reports with multi-axis display for temperature, humidity, CO2, and light; customer parameters stay in config files to avoid hardcoding sensitive values.
• Message delivery: Can send images/Markdown through DingTalk, including scheduled collection and push workflows.

Boundaries and notes

The skill depends on JJR IoT platform credentials and is not suitable for generic IoT environments without those permissions. Query results depend on correct productKey, deviceName, time range, and device online status. Image downloads may fail due to OSS permission issues or expired URLs. Keep client_secret secure, use isolated config per customer, and do not commit secrets to version control.