How do I install the Analyze Trace skill?

Run npx skillmds@latest add tomevault-io/analyze-trace in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Analyze Trace safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Analyze Trace?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Analyze Trace free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Analyze Trace?

tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Analyze Trace by tomevault-io — Agent Skill

Name: Analyze Trace
Author: tomevault-io

Analyze Trace

Walk a Lensflare trace span-by-span and explain timing, errors, and parent/child structure. Use when the user pastes a trace id, asks "why is this slow", or wants a request-level post-mortem of a real production / dev request captured in Lensflare. Use when this capability is needed.

tomevault-io Updated Sep 17, 2026

File contents

Analyze a trace in Lensflare

If the user supplied a traceId, use it. Otherwise call lensflare:queryTelemetry to find a relevant trace (filter by serviceName, time, or attributes). Pick the trace whose root span has the highest durationUs or whose status = "error".
Call lensflare:getTrace with the resolved datasetId and traceId. If it returns null, double-check the traceId with queryTelemetry (traceId = "<id>") before reporting back.
Read summary.spanCount and summary.erroredSpanCount. If errors exist, lead with them: name + service + durationUs.
Walk spans in order. Highlight the longest span, the deepest chain, and any span whose status = "error" or whose duration is more than 50% of its parent's.
Surface relatedEvents only when they materially explain a failure (e.g. an exception event on the failing span).
End with a one-line root-cause hypothesis and a suggested follow-up query. Example: "to see all requests with this error: attributes.http.status_code = 500 and serviceName = \"api\"".

Source: voidhashcom/lensflare — distributed by TomeVault.

Frequently asked questions