Skip to content

feat(ai): opt-in AI run summary on the HTML dashboard - #168

Merged
shossain786 merged 1 commit into
masterfrom
feat/ai-run-summary
Oct 18, 2026
Merged

shossain786 merged 1 commit into
masterfrom
feat/ai-run-summary

Conversation

@shossain786

Copy link
Copy Markdown
Contributor

Adds ai.runSummary (default off): one API call at suite end that summarises the run's failures grouped by cause, shown as a card at the top of the dashboard and exported as aiRunSummary.

  • New AiRunSummarizer, reusing AiFailureAnalyzer's API call, key resolution and timeout. Any failure returns null; the suite result is never affected.
  • Data sent: run totals plus, for the 8 largest same-cause groups, category, first 300 chars of the error message and up to 5 test names. No stack traces, steps or artifacts. Documented in reporting/html-report.md.
  • Output is HTML-escaped in the report.
  • Tests: prompt grouping/ordering/caps, disabled-by-default, report rendering + escaping. mvn clean verify: 730 tests, 0 failures.
  • Not verified: a live call to the Claude API (no key used); the request path is the unchanged one used by failure analysis.

🤖 Generated with Claude Code

ai.runSummary groups the run's failures by cause and asks Claude for a
short 'what to look at first' paragraph, stored as aiRunSummary in the
metrics JSON and shown at the top of the dashboard. Disabled by default;
never affects the suite result.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@shossain786
shossain786 merged commit bf85774 into master Oct 18, 2026
2 checks passed
@shossain786
shossain786 deleted the feat/ai-run-summary branch October 18, 2026 02:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant