Skip to main content
hc
Submit
agentsSource-backedReview first Safety ✓ Privacy ✓

OpenTelemetry Trace Analysis Agent

Community reusable agent prompt for analyzing OpenTelemetry traces from agent hosts using official trace signal documentation: span hierarchy review, latency attribution, error propagation, and privacy-safe instrumentation recommendations.

by kiannidev·added 2026-06-16·
Claude Code
HarnessClaude Code
Review first — review before installing

Open the source and read safety notes before installing.

Safety notes

  • Do not copy raw span attributes containing prompts into public tickets.
  • Recommend redaction and hashing when traces store user content.
  • High-cardinality attributes can explode observability cost.
  • Trace analysis informs changes; validate mitigations in staging before production.

Privacy notes

  • Span attributes may accidentally capture API keys or JWT fragments if developers log headers.
  • Shared trace viewers must restrict access when spans include proprietary code paths.
  • Prefer aggregate latency reports over attaching full trace exports externally.

Prerequisites

  • Exported trace JSON or APM links for failing agent sessions.
  • Service names for model client, tool executor, and storage layers.
  • Instrumentation policy on whether prompt content may appear in span attributes.
  • Baseline latency expectations for model turns and top MCP tools.

Schema details

Install type
copy
Troubleshooting
No
Source repository stats
Scope
Source repo
Full copyable content
## Content

OpenTelemetry Trace Analysis Agent is a community-authored reusable prompt for reviewing
traces from agent systems. It applies official OpenTelemetry trace signal documentation—not
a vendor APM support agent.

## Scope Note

This prompt operationalizes OpenTelemetry trace concepts from opentelemetry.io. Host-specific
instrumentation details remain with your platform team.

## Agent Prompt

You are an OpenTelemetry trace analyst for agent systems. Explain traces without leaking
prompt content, using official trace signal documentation.

Workflow:

1. **Trace selection.** Pick representative failed and slow traces with complete span trees.
2. **Hierarchy.** Map root session spans to model calls, tool executions, subagents, and I/O.
3. **Latency attribution.** Quantify time in model API, MCP tools, DB, and queue waits.
4. **Errors.** Follow exception events and status codes across service boundaries.
5. **Gaps.** Note missing spans for MCP calls or broken context propagation.
6. **Privacy.** Flag spans logging prompts, diffs, or secrets; recommend redaction patterns.
7. **Recommendations.** Suggest span naming, attributes, sampling, and dashboards.

Output contract:

- Critical path summary with percent latency breakdown.
- Error chain with service owners.
- Instrumentation and privacy findings.
- Prioritized engineering fixes.

## Features

- Applies OpenTelemetry trace docs to agent-specific span patterns.
- Separates model latency from MCP dependency latency.
- Promotes privacy-safe attribute design.
- Produces engineer-ready instrumentation backlogs.

## Use Cases

- Debug p95 latency regressions after adding a new MCP server.
- Audit traces before enabling enterprise content logging policies.
- Validate OpenTelemetry export during Agent SDK hardening.
- Support post-incident reviews with evidence-backed span analysis.

## Source Notes

Verified against OpenTelemetry trace signal documentation on **2026-06-16**:

- Official docs describe traces as trees of spans representing causal relationships between
  operations in distributed systems.
- Trace documentation covers span attributes, events, and status fields used to diagnose latency
  and errors across services.
- OpenTelemetry guidance emphasizes sampling and cardinality control when instrumenting
  high-volume agent workloads.

## Duplicate Check

Checked content/agents and content/guides for trace analysis coverage.
llm-agent-application-observability is a guides entry for general OTel setup.
No agents entry provides trace-structure analysis with privacy-safe instrumentation
recommendations grounded in official OpenTelemetry trace signal documentation.

## Editorial Disclosure

Submitted as an independent community agent entry by kiannidev, based on public
OpenTelemetry documentation and the public OpenTelemetry specification repository.
No paid placement, referral, or affiliate relationship.

## Sources

- OpenTelemetry traces - https://opentelemetry.io/docs/concepts/signals/traces/
- OpenTelemetry observability primer - https://opentelemetry.io/docs/concepts/observability-primer/
- OpenTelemetry specification repository - https://github.com/open-telemetry/opentelemetry-specification

About this resource

Content

OpenTelemetry Trace Analysis Agent is a community-authored reusable prompt for reviewing traces from agent systems. It applies official OpenTelemetry trace signal documentation—not a vendor APM support agent.

Scope Note

This prompt operationalizes OpenTelemetry trace concepts from opentelemetry.io. Host-specific instrumentation details remain with your platform team.

Agent Prompt

You are an OpenTelemetry trace analyst for agent systems. Explain traces without leaking prompt content, using official trace signal documentation.

Workflow:

  1. Trace selection. Pick representative failed and slow traces with complete span trees.
  2. Hierarchy. Map root session spans to model calls, tool executions, subagents, and I/O.
  3. Latency attribution. Quantify time in model API, MCP tools, DB, and queue waits.
  4. Errors. Follow exception events and status codes across service boundaries.
  5. Gaps. Note missing spans for MCP calls or broken context propagation.
  6. Privacy. Flag spans logging prompts, diffs, or secrets; recommend redaction patterns.
  7. Recommendations. Suggest span naming, attributes, sampling, and dashboards.

Output contract:

  • Critical path summary with percent latency breakdown.
  • Error chain with service owners.
  • Instrumentation and privacy findings.
  • Prioritized engineering fixes.

Features

  • Applies OpenTelemetry trace docs to agent-specific span patterns.
  • Separates model latency from MCP dependency latency.
  • Promotes privacy-safe attribute design.
  • Produces engineer-ready instrumentation backlogs.

Use Cases

  • Debug p95 latency regressions after adding a new MCP server.
  • Audit traces before enabling enterprise content logging policies.
  • Validate OpenTelemetry export during Agent SDK hardening.
  • Support post-incident reviews with evidence-backed span analysis.

Source Notes

Verified against OpenTelemetry trace signal documentation on 2026-06-16:

  • Official docs describe traces as trees of spans representing causal relationships between operations in distributed systems.
  • Trace documentation covers span attributes, events, and status fields used to diagnose latency and errors across services.
  • OpenTelemetry guidance emphasizes sampling and cardinality control when instrumenting high-volume agent workloads.

Duplicate Check

Checked content/agents and content/guides for trace analysis coverage. llm-agent-application-observability is a guides entry for general OTel setup. No agents entry provides trace-structure analysis with privacy-safe instrumentation recommendations grounded in official OpenTelemetry trace signal documentation.

Editorial Disclosure

Submitted as an independent community agent entry by kiannidev, based on public OpenTelemetry documentation and the public OpenTelemetry specification repository. No paid placement, referral, or affiliate relationship.

Sources

#opentelemetry#traces#observability#agents#sre

Source citations

Add this badge to your README

Show that OpenTelemetry Trace Analysis Agent is listed on HeyClaude. Paste this Markdown into your README — it renders the badge and links back to this page.

Listed on HeyClaude
[![Listed on HeyClaude](https://heyclau.de/badge/agents/opentelemetry-trace-analysis-agent.svg)](https://heyclau.de/entry/agents/opentelemetry-trace-analysis-agent)

How it compares

OpenTelemetry Trace Analysis Agent side by side with its closest alternative on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.

FieldOpenTelemetry Trace Analysis Agent

Community reusable agent prompt for analyzing OpenTelemetry traces from agent hosts using official trace signal documentation: span hierarchy review, latency attribution, error propagation, and privacy-safe instrumentation recommendations.

Open dossier 		Agent Observability SRE Agent

Community reusable agent prompt for Claude Code analytics and agent platform on-call using official analytics documentation: usage signals, session failure triage, MCP latency patterns, and SRE runbooks for agent hosting teams.

Open dossier
Trust
Install riskReview firstReview first
Notes Safety ✓ Privacy ✓ Safety ✓ Privacy ✓
Categoryagentsagents
Sourcesource-backedsource-backed
Authorkiannidevkiannidev
Added2026-06-162026-06-16
Platforms
Claude Code
Claude Code
Source repo——
Safety notes✓Do not copy raw span attributes containing prompts into public tickets. Recommend redaction and hashing when traces store user content. High-cardinality attributes can explode observability cost. Trace analysis informs changes; validate mitigations in staging before production.✓Incident commands must not exfiltrate customer prompts into public tickets. Scaling replicas without reviewing tool side effects can amplify destructive MCP calls. Disabling tracing to reduce noise may hide regressions—prefer sampling over full off. Rollback plans should include MCP allowlist and permission settings, not only code.
Privacy notes✓Span attributes may accidentally capture API keys or JWT fragments if developers log headers. Shared trace viewers must restrict access when spans include proprietary code paths. Prefer aggregate latency reports over attaching full trace exports externally.✓Analytics and logs may contain prompts, diffs, and credentials if misconfigured. Recommend redaction before exporting incident timelines externally. Shared dashboards should aggregate metrics without raw user content fields.
Prerequisites
  • Exported trace JSON or APM links for failing agent sessions.
  • Service names for model client, tool executor, and storage layers.
  • Instrumentation policy on whether prompt content may appear in span attributes.
  • Baseline latency expectations for model turns and top MCP tools.
  • Access to Claude Code analytics or org usage exports for affected teams.
  • Logs from agent hosts, MCP gateways, and background workers when self-hosting SDK workloads.
  • Defined SLOs for session completion time and error budgets for agent tasks.
  • Architecture diagram showing model calls, tool execution, and persistence layers.
Install——
Config——
Citations
ClaimUnclaimedUnclaimed

Signals

Loading live community signals…

More like this, weekly

A short, calm digest of reviewed Claude resources. Unsubscribe any time.