Agent skill · data analytics · jeremylongshore
salesforce-advanced-troubleshooting
Analyze complex Salesforce integration failures across request traces, metadata, query plans, automation, locks, jobs, events, and limits. Use when basic triage stalls. Trigger with "deep diagnose Salesforce".
Why this skill is useful
Provides advanced debugging techniques and specific commands for Salesforce performance issues that the AI wouldn't reliably generate on its own.
What it needs
Requires salesforce-cli installed locally. Requires salesforce account access. About 2k tokens when loaded. Last updated 2026-08-07. 2,604 stars on the source repository.
What this skill does
Salesforce Advanced Integration Troubleshooting Overview Build and falsify layered hypotheses with minimum evidence rather than changing several org, query, permission, or retry variables at once. Prerequisites Basic triage record, exact time window, environment, release, principal, request or job IDs, and current symptom Architecture, API and event contracts, object metadata, automation map, limits, deployments, and prior baselines Incident commander, application, Salesforce platform, security, data, and vendor support owners Tool Discipline Use Read, Glob, and Grep to inspect approved repository and evidence files, WebFetch to re-check current first-party Salesforce documentation, and Write or Edit only for secretless plans, fixtures, configuration, and redacted receipts. Current Contract Salesforce diagnostic behavior spans versioned REST responses, debug logs, query feedback, async jobs, event usage, and org configuration. REST query explain is Beta and must be corroborated; diagnostic surface availability varies by entitlement and permission. Authentication Use the narrowest diagnostic principal and time window. Never enable broad logging or collect unredacted queries, records, tokens, sessions, org identifiers, or user data without explicit authorization. Instructions 1. Restate the invariant, symptom, first failure, blast radius, changes, retry history, and what basic triage already ruled out. 2. Construct hypotheses across client, network, OAuth, permission, sharing, metadata, query, automation, lock, limit, async, event, and downstream layers. 3. Collect one discriminating redacted signal per hypothesis using request IDs, logs, limits, job results, metadata fingerprints, and deployment IDs. 4. Use query explain or plan evidence only under its documented status and compare it with representative measured execution. 5. Reproduce with synthetic data or read-only requests, then change one variable at a time under a defined stop condition. 6. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills jeremylongshore/salesforce-advanced-troubleshooting