Observe Trace
- 644 installs
- 67k repo stars
- Updated August 4, 2026
- ruvnet/ruflo
observe-trace is a ruflo Claude skill that collects distributed trace spans and builds a trace tree for developers who need to replay agent reasoning steps and tool calls across coordinated tasks.
About
observe-trace is a distributed tracing skill in ruvnet/ruflo that takes a task-id argument and collects spans via claude-flow MCP tools including agentdb_semantic-route and agentdb_context-synthesize. It builds a visual trace tree showing execution flow, per-span timing, bottlenecks, and how agents coordinated during a task. Developers reach for observe-trace when they need to understand which spans ran, how long each took, and where coordination failed during multi-agent execution.
- Real-time observation of agent traces and decision paths
- Captures tool calls, LLM responses, and intermediate reasoning
- Supports replay and debugging of complex agent workflows
- Lightweight integration with existing Claude, Cursor, and Codex agents
- Helps identify why an agent deviated from expected behavior
Observe Trace by the numbers
- 644 all-time installs (skills.sh)
- +6 installs in the week ending Aug 5, 2026 (Skillselion tracking)
- Ranked #1,519 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/ruvnet/ruflo --skill observe-traceAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 644 |
|---|---|
| repo stars | ★ 67k |
| Last updated | August 4, 2026 |
| Repository | ruvnet/ruflo ↗ |
How do you trace multi-agent task execution spans?
Monitor, log, and replay agent reasoning steps and tool calls in real time.
Who is it for?
Developers debugging ruflo or claude-flow multi-agent tasks who need span-level timing and coordination visibility.
Skip if: Aggregate health snapshots across time periods—use observe-metrics instead.
When should I use this skill?
The user provides a task-id or asks how a multi-agent task executed, which tools ran, or where bottlenecks occurred.
What you get
Visual trace tree with span timing, execution flow, and bottleneck identification for a task-id.
- Trace tree visualization
- Span timing breakdown
- Bottleneck report
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.
Steps
1. Collect spans -- call mcp__claude-flow__memory_search --namespace observability (or memory_list) to retrieve all spans matching the <task-id>. The memory_* tool family routes by namespace; agentdb_hierarchical-* does NOT (it routes by tier working|episodic|semantic), so use memory_* here. See ruflo-agentdb ADR-0001 §"Namespace convention". 2. Build trace tree -- organize spans into a parent-child hierarchy using parentSpanId references, with the root span at the top 3. Calculate timing -- for each span, compute duration (endTime - startTime), and identify the critical path (longest chain of sequential spans) 4. Identify bottlenecks -- flag spans where duration exceeds the p95 for that operation type, or where gaps between spans suggest idle time 5. Synthesize -- call mcp__claude-flow__agentdb_context-synthesize to combine span metadata into a narrative summary of the execution flow 6. Report -- display the trace tree with: span name, agent, duration, status (OK/ERROR), and bottleneck flag; include total trace duration and critical path duration
CLI alternative
npx @claude-flow/cli@latest memory search --query "trace spans for task TASK_ID" --namespace observabilityRelated skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la495k196k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.402k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.391k94.7k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.386k16k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k10FAQ
What input does observe-trace require?
observe-trace takes a task-id argument, collects distributed spans for that task via claude-flow MCP, and assembles a trace tree showing execution flow and per-span timing.
How is observe-trace different from observe-metrics?
observe-trace drills into one task-id with span-level trace trees and bottlenecks; observe-metrics aggregates counters, gauges, and histograms across a time period for swarm health.