AGENT SKILL

exa-advanced-troubleshooting

Apply Exa advanced debugging techniques for hard-to-diagnose issues. Use when standard troubleshooting fails, investigating complex race conditions, or preparing evidence bundles for Exa support escalation. Trigger with phrases like "exa hard bug", "exa mystery error", "exa impossible to debug", "difficult exa issue", "exa deep debug".

适用职业开发工程师
License 许可
MIT
最近更新
2026-11-15
Stars
★ 2.0k
Forks
233

01 前言 Front Matter

name
exa-advanced-troubleshooting
author
Jeremy Longshore <jeremy@intentsolutions.io>
license
MIT
version
1.0.0
description
Apply Exa advanced debugging techniques for hard-to-diagnose issues. Use when standard troubleshooting fails, investigating complex race conditions, or preparing evidence bundles for Exa support escalation. Trigger with phrases like "exa hard bug", "exa mystery error", "exa impossible to debug", "difficult exa issue", "exa deep debug".
allowed-tools
Read, Grep, Bash(kubectl:*), Bash(curl:*), Bash(tcpdump:*)

03 SKILL.md 正文

Exa Advanced Troubleshooting

Overview

Deep debugging techniques for complex Exa issues that resist standard troubleshooting.

Prerequisites

  • Access to production logs and metrics
  • kubectl access to clusters
  • Network capture tools available
  • Understanding of distributed tracing

Evidence Collection Framework

Comprehensive Debug Bundle

#!/bin/bash
# advanced-exa-debug.sh

BUNDLE="exa-advanced-debug-$(date +%Y%m%d-%H%M%S)"
mkdir -p "$BUNDLE"/{logs,metrics,network,config,traces}

# 1. Extended logs (1 hour window)
kubectl logs -l app=exa-integration --since=1h > "$BUNDLE/logs/pods.log"
journalctl -u exa-service --since "1 hour ago" > "$BUNDLE/logs/system.log"

# 2. Metrics dump
curl -s localhost:9090/api/v1/query?query=exa_requests_total > "$BUNDLE/metrics/requests.json"
curl -s localhost:9090/api/v1/query?query=exa_errors_total > "$BUNDLE/metrics/errors.json"

# 3. Network capture (30 seconds)
timeout 30 tcpdump -i any port 443 -w "$BUNDLE/network/capture.pcap" &

# 4. Distributed traces
curl -s localhost:16686/api/traces?service=exa > "$BUNDLE/traces/jaeger.json"

# 5. Configuration state
kubectl get cm exa-config -o yaml > "$BUNDLE/config/configmap.yaml"
kubectl get secret exa-secrets -o yaml > "$BUNDLE/config/secrets-redacted.yaml"

tar -czf "$BUNDLE.tar.gz" "$BUNDLE"
echo "Advanced debug bundle: $BUNDLE.tar.gz"

Systematic Isolation

Layer-by-Layer Testing

// Test each layer independently
async function diagnoseExaIssue(): Promise<DiagnosisReport> {
  const results: DiagnosisResult[] = [];

  // Layer 1: Network connectivity
  results.push(await testNetworkConnectivity());

  // Layer 2: DNS resolution
  results.push(await testDNSResolution('api.exa.com'));

  // Layer 3: TLS handshake
  results.push(await testTLSHandshake('api.exa.com'));

  // Layer 4: Authentication
  results.push(await testAuthentication());

  // Layer 5: API response
  results.push(await testAPIResponse());

  // Layer 6: Response parsing
  results.push(await testResponseParsing());

  return { results, firstFailure: results.find(r => !r.success) };
}

Minimal Reproduction

// Strip down to absolute minimum
async function minimalRepro(): Promise<void> {
  // 1. Fresh client, no customization
  const client = new ExaClient({
    apiKey: process.env.EXA_API_KEY!,
  });

  // 2. Simplest possible call
  try {
    const result = await client.ping();
    console.log('Ping successful:', result);
  } catch (error) {
    console.error('Ping failed:', {
      message: error.message,
      code: error.code,
      stack: error.stack,
    });
  }
}

Timing Analysis

class TimingAnalyzer {
  private timings: Map<string, number[]> = new Map();

  async measure<T>(label: string, fn: () => Promise<T>): Promise<T> {
    const start = performance.now();
    try {
      return await fn();
    } finally {
      const duration = performance.now() - start;
      const existing = this.timings.get(label) || [];
      existing.push(duration);
      this.timings.set(label, existing);
    }
  }

  report(): TimingReport {
    const report: TimingReport = {};
    for (const [label, times] of this.timings) {
      report[label] = {
        count: times.length,
        min: Math.min(...times),
        max: Math.max(...times),
        avg: times.reduce((a, b) => a + b, 0) / times.length,
        p95: this.percentile(times, 95),
      };
    }
    return report;
  }
}

Memory and Resource Analysis

// Detect memory leaks in Exa client usage
const heapUsed: number[] = [];

setInterval(() => {
  const usage = process.memoryUsage();
  heapUsed.push(usage.heapUsed);

  // Alert on sustained growth
  if (heapUsed.length > 60) { // 1 hour at 1/min
    const trend = heapUsed[59] - heapUsed[0];
    if (trend > 100 * 1024 * 1024) { // 100MB growth
      console.warn('Potential memory leak in exa integration');
    }
  }
}, 60000);

Race Condition Detection

// Detect concurrent access issues
class ExaConcurrencyChecker {
  private inProgress: Set<string> = new Set();

  async execute<T>(key: string, fn: () => Promise<T>): Promise<T> {
    if (this.inProgress.has(key)) {
      console.warn(`Concurrent access detected for ${key}`);
    }

    this.inProgress.add(key);
    try {
      return await fn();
    } finally {
      this.inProgress.delete(key);
    }
  }
}

Support Escalation Template

## Exa Support Escalation

**Severity:** P[1-4]
**Request ID:** [from error response]
**Timestamp:** [ISO 8601]

### Issue Summary
[One paragraph description]

### Steps to Reproduce
1. [Step 1]
2. [Step 2]

### Expected vs Actual
- Expected: [behavior]
- Actual: [behavior]

### Evidence Attached
- [ ] Debug bundle (exa-advanced-debug-*.tar.gz)
- [ ] Minimal reproduction code
- [ ] Timing analysis
- [ ] Network capture (if relevant)

### Workarounds Attempted
1. [Workaround 1] - Result: [outcome]
2. [Workaround 2] - Result: [outcome]

Instructions

Step 1: Collect Evidence Bundle

Run the comprehensive debug script to gather all relevant data.

Step 2: Systematic Isolation

Test each layer independently to identify the failure point.

Step 3: Create Minimal Reproduction

Strip down to the simplest failing case.

Step 4: Escalate with Evidence

Use the support template with all collected evidence.

Output

  • Comprehensive debug bundle collected
  • Failure layer identified
  • Minimal reproduction created
  • Support escalation submitted

Error Handling

Issue Cause Solution
Can't reproduce Race condition Add timing analysis
Intermittent failure Timing-dependent Increase sample size
No useful logs Missing instrumentation Add debug logging
Memory growth Resource leak Use heap profiling

Examples

Quick Layer Test

# Test each layer in sequence
curl -v https://api.exa.com/health 2>&1 | grep -E "(Connected|TLS|HTTP)"

Resources

Next Steps

For load testing, see exa-load-scale.

04 同仓库更多 Skills

Dicklesworthstone
Dicklesworthstone/pi_agent_rust

theme-factory

Toolkit for styling artifacts with a theme. These artifacts can be slides, docs, reportings, HTML landing pages, etc. There are 10 pre-set themes with colors/fonts that you can apply to any artifact that has been creating, or can generate a new theme on-the-fly.

视觉设计
★ 2.0k
Dicklesworthstone
Dicklesworthstone/pi_agent_rust

web-artifacts-builder

Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui). Use for complex artifacts requiring state management, routing, or shadcn/ui components - not for simple single-file HTML/JSX artifacts.

前端开发
★ 2.0k
Dicklesworthstone
Dicklesworthstone/pi_agent_rust

webapp-testing

Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.

测试与质量前端开发
★ 2.0k
Dicklesworthstone
Dicklesworthstone/pi_agent_rust

neuropixels-analysis

Neuropixels neural recording analysis. Load SpikeGLX/OpenEphys data, preprocess, motion correction, Kilosort4 spike sorting, quality metrics, Allen/IBL curation, AI-assisted visual analysis, for Neuropixels 1.0/2.0 extracellular electrophysiology. Use when working with neural recordings, spike sorting, extracellular electrophysiology, or when the user mentions Neuropixels, SpikeGLX, Open Ephys, Kilosort, quality metrics, or unit curation.

★ 2.0k
Dicklesworthstone
Dicklesworthstone/pi_agent_rust

skill-creator

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

Skill与智能体开发
★ 2.0k
Dicklesworthstone
Dicklesworthstone/pi_agent_rust

notebooklm

Use this skill to query your Google NotebookLM notebooks directly from Claude Code for source-grounded, citation-backed answers from Gemini. Browser automation, library management, persistent auth. Drastically reduced hallucinations through document-only responses.

Skill与智能体开发
★ 2.0k