Dev Log Monitoring FAQs

Question 1

When should I use this skill over standard debugging skills?

Accepted Answer

You should use this skill specifically for testing ScreenGraph end-to-end flows, such as drift detection and agent runs. It is the best choice when you need to observe how multiple services interact, verify WebSocket event streaming, or analyze the state machine transitions of an autonomous agent.

Question 2

What kind of data is included in the final summary report?

Accepted Answer

The skill generates a comprehensive summary including Run IDs, app metadata, a step-by-step agent state machine timeline, graph projection results (like discovered screens), WebSocket backfill counts, and specific frontend console or network errors.

Question 3

What is the Dev Log Monitoring Claude Code skill?

Accepted Answer

The Dev Log Monitoring skill is a specialized capability for Claude Code designed to capture, pipe, and analyze real-time logs from both backend (Encore.ts) and frontend (SvelteKit) services. It allows Claude to observe system behavior during complex end-to-end tests and agent executions.

Question 4

How does this skill improve the debugging workflow?

Accepted Answer

It automates the tedious process of manual log tailing and correlation. By piping logs to temporary files and using specific grep patterns, the skill provides a systematic timeline of events, including node transitions, event recording sequences, and performance metrics, all within the Claude Code interface.

Question 5

Can this skill interact with browser-based applications?

Accepted Answer

Yes, it is designed to work in tandem with Playwright MCP. This allows Claude to trigger frontend actions (like clicking buttons or navigating pages) while simultaneously monitoring the backend logs to see how those actions trigger drift detection or graph projection events.

Dev Log Monitoring

Dev Log Monitoring

Key Features

Use Cases

Key Features

Use Cases