Observe Trace

You need to see how a distributed task actually executed across multiple agents, and this builds you a trace tree from span data stored in observability memory. It pulls all the spans for a task ID, reconstructs the parent-child hierarchy, calculates durations, and flags bottlenecks where things got slow or agents sat idle. The synthesis step is smart: it uses agentdb to turn raw span metadata into a readable narrative of what happened. Most useful when debugging coordination problems or hunting performance issues in multi-agent workflows. The trace tree shows you the critical path and total duration, so you can see exactly where time got burned.

skills.sh installs: 733
GitHub stars: 81.4K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add ruvnet/ruflo --skill observe-trace --agent claude-code

Installs into .claude/skills of the current project.

Files

Observe Trace

Collect distributed trace spans for a task and build a visual trace tree showing the execution flow, timing, and bottlenecks.

When to use

When you need to understand how a task was executed across agents -- which spans ran, how long each took, where bottlenecks occurred, and how agents coordinated.