From hubspot-pack
Debug complex HubSpot API issues with systematic isolation and evidence collection. Use when standard troubleshooting fails, investigating intermittent CRM errors, or preparing evidence bundles for HubSpot support escalation. Trigger with phrases like "hubspot hard bug", "hubspot mystery error", "hubspot intermittent failure", "hubspot deep debug", "hubspot support ticket".
How this skill is triggered — by the user, by Claude, or both
Slash command
/hubspot-pack:hubspot-advanced-troubleshootingThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Deep debugging techniques for complex HubSpot API issues: systematic layer testing, timing analysis, correlation ID tracking, and support escalation.
Deep debugging techniques for complex HubSpot API issues: systematic layer testing, timing analysis, correlation ID tracking, and support escalation.
curl and jq availableTest each layer independently to isolate the failure:
#!/bin/bash
# hubspot-layer-test.sh
echo "=== HubSpot Layer-by-Layer Diagnostic ==="
# Layer 1: DNS Resolution
echo "1. DNS Resolution"
dig api.hubapi.com +short || echo "FAIL: DNS resolution"
# Layer 2: TCP Connectivity
echo "2. TCP Connectivity"
timeout 5 bash -c 'echo > /dev/tcp/api.hubapi.com/443' 2>/dev/null \
&& echo "OK" || echo "FAIL: Cannot reach port 443"
# Layer 3: TLS Handshake
echo "3. TLS Handshake"
echo | openssl s_client -connect api.hubapi.com:443 2>/dev/null | grep "Verify return code"
# Layer 4: HTTP Response (unauthenticated)
echo "4. HTTP Response (no auth)"
curl -so /dev/null -w "HTTP %{http_code} in %{time_total}s\n" \
https://api.hubapi.com/crm/v3/objects/contacts?limit=1
# Layer 5: Authenticated Request
echo "5. Authenticated Request"
RESPONSE=$(curl -s -w "\n%{http_code}" \
https://api.hubapi.com/crm/v3/objects/contacts?limit=1 \
-H "Authorization: Bearer $HUBSPOT_ACCESS_TOKEN")
HTTP_CODE=$(echo "$RESPONSE" | tail -1)
BODY=$(echo "$RESPONSE" | head -n -1)
echo "HTTP $HTTP_CODE"
if [ "$HTTP_CODE" = "200" ]; then
echo "OK: API accessible"
echo "Records: $(echo $BODY | jq '.results | length')"
else
echo "FAIL: $(echo $BODY | jq -r '.category // .message')"
echo "Correlation ID: $(echo $BODY | jq -r '.correlationId // "none"')"
fi
# Layer 6: Rate Limit State
echo "6. Rate Limit State"
curl -sI https://api.hubapi.com/crm/v3/objects/contacts?limit=1 \
-H "Authorization: Bearer $HUBSPOT_ACCESS_TOKEN" \
| grep -i "ratelimit" | sed 's/^/ /'
Every HubSpot error response includes a correlationId. Track these for support:
import * as hubspot from '@hubspot/api-client';
// Collect correlation IDs from errors
const errorLog: Array<{
timestamp: string;
correlationId: string;
statusCode: number;
category: string;
message: string;
endpoint: string;
}> = [];
async function debuggedApiCall<T>(
endpoint: string,
operation: () => Promise<T>
): Promise<T> {
try {
return await operation();
} catch (error: any) {
const body = error?.body || {};
const entry = {
timestamp: new Date().toISOString(),
correlationId: body.correlationId || 'unknown',
statusCode: error?.code || error?.statusCode || 500,
category: body.category || 'UNKNOWN',
message: body.message || error.message,
endpoint,
};
errorLog.push(entry);
console.error('HubSpot error:', entry);
throw error;
}
}
// Dump error log for support ticket
function getErrorReport(): string {
return errorLog.map(e =>
`[${e.timestamp}] ${e.statusCode} ${e.category} | ${e.correlationId} | ${e.endpoint}: ${e.message}`
).join('\n');
}
async function timingAnalysis() {
const operations = [
{
name: 'GET contacts (single)',
fn: () => client.crm.contacts.basicApi.getPage(1, undefined, ['email']),
},
{
name: 'SEARCH contacts',
fn: () => client.crm.contacts.searchApi.doSearch({
filterGroups: [{
filters: [{ propertyName: 'lifecyclestage', operator: 'EQ', value: 'lead' }],
}],
properties: ['email'], limit: 1, after: 0, sorts: [],
}),
},
{
name: 'GET pipelines',
fn: () => client.crm.pipelines.pipelinesApi.getAll('deals'),
},
{
name: 'GET properties',
fn: () => client.crm.properties.coreApi.getAll('contacts'),
},
];
console.log('=== HubSpot Timing Analysis ===');
for (const op of operations) {
const times: number[] = [];
for (let i = 0; i < 5; i++) {
const start = performance.now();
try {
await op.fn();
times.push(performance.now() - start);
} catch {
times.push(-1); // mark failures
}
}
const successful = times.filter(t => t >= 0);
if (successful.length > 0) {
console.log(`${op.name}:
Min: ${Math.min(...successful).toFixed(0)}ms
Max: ${Math.max(...successful).toFixed(0)}ms
Avg: ${(successful.reduce((a, b) => a + b, 0) / successful.length).toFixed(0)}ms
Failures: ${times.length - successful.length}/5`);
} else {
console.log(`${op.name}: ALL FAILED`);
}
}
}
// Reproduce common intermittent issues
// Test 1: Rapid sequential calls (rate limit sensitivity)
async function testRateLimit(requestCount = 20) {
console.log(`Sending ${requestCount} rapid requests...`);
let success = 0, rateLimited = 0, errors = 0;
for (let i = 0; i < requestCount; i++) {
try {
await client.crm.contacts.basicApi.getPage(1, undefined, ['email']);
success++;
} catch (error: any) {
if (error?.code === 429) rateLimited++;
else errors++;
}
}
console.log(`Results: ${success} ok, ${rateLimited} rate limited, ${errors} errors`);
}
// Test 2: Concurrent requests (connection pool issues)
async function testConcurrency(concurrent = 10) {
console.log(`Sending ${concurrent} concurrent requests...`);
const results = await Promise.allSettled(
Array.from({ length: concurrent }, () =>
client.crm.contacts.basicApi.getPage(1, undefined, ['email'])
)
);
const fulfilled = results.filter(r => r.status === 'fulfilled').length;
const rejected = results.filter(r => r.status === 'rejected').length;
console.log(`Results: ${fulfilled} ok, ${rejected} failed`);
const failures = results.filter(r => r.status === 'rejected') as PromiseRejectedResult[];
for (const f of failures) {
console.log(` Error: ${f.reason?.code} ${f.reason?.body?.category}`);
}
}
## HubSpot Support Escalation
**Severity:** P[1-4]
**Portal ID:** [your portal ID]
**Correlation IDs:** [list from error log]
### Issue Summary
[1-2 sentence description]
### Steps to Reproduce
1. Make API call: POST /crm/v3/objects/contacts/search
2. With body: { "filterGroups": [...], "properties": ["email"], "limit": 10 }
3. Observe: 500 Internal Server Error
### Error Response
```json
{
"status": "error",
"message": "[exact error message]",
"correlationId": "[exact ID]",
"category": "[exact category]"
}
## Output
- Layer-by-layer diagnostic isolating the failure point
- Correlation ID collection for support tickets
- Timing analysis identifying latency patterns
- Reproduction scripts for rate limit and concurrency issues
- Support escalation template with all required evidence
## Error Handling
| Scenario | Diagnostic | Next Step |
|----------|-----------|-----------|
| DNS fails | Check resolver config | Try `8.8.8.8` resolver |
| TLS fails | Certificate issue | Check system CA bundle |
| 401 after working | Token rotated | Regenerate in Settings |
| Intermittent 500 | HubSpot server issue | Collect correlation IDs, file ticket |
| Latency spikes | Network or HubSpot load | Run timing analysis over time |
## Resources
- [HubSpot Support Portal](https://help.hubspot.com/)
- [HubSpot Developer Community](https://community.hubspot.com/t5/APIs-Integrations/bd-p/integration)
- [Error Handling Guide](https://developers.hubspot.com/docs/api-reference/error-handling)
## Next Steps
For load testing, see `hubspot-load-scale`.
npx claudepluginhub flight505/skill-forge --plugin hubspot-packCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.