afv-library/skills/platform-apex-logs-debug/references/analysis-playbook.md

1.3 KiB

Debug Analysis Playbook

Use this playbook when platform-apex-logs-debug is active and you need the expanded workflow.

1. Gather context

Collect:

  • org alias
  • failing transaction or test context
  • approximate time window
  • relevant user / record / request identifiers

2. Retrieve logs

Preferred commands:

sf apex list log --target-org <alias> --json
sf apex get log --log-id <id> --target-org <alias>
sf apex tail log --target-org <alias> --color

See cli-commands.md for more options.

3. Analyze in this order

  1. transaction entry point
  2. exceptions and fatal errors
  3. governor limits
  4. SOQL / DML repetition patterns
  5. CPU / heap hotspots
  6. callout timing and external failures

4. Classify severity

  • Critical — runtime failure, hard limit, data corruption risk
  • Warning — near-limit, non-selective query, slow path
  • Info — optimization opportunity, cleanup item, observability gap

5. Propose fixes

Prefer fixes that are:

  • root-cause oriented
  • bulk-safe
  • testable
  • deployable in one clean change set

6. Loop with adjacent skills

  • use sf-apex for code fixes
  • use platform-apex-test-run to reproduce and verify
  • use platform-metadata-deploy to deploy fixes
  • use platform-data-manage when the issue depends on missing or malformed test data