Simulated incident story
This demo data is read-only and follows one deploy from failing tests to recovery.
- Bad deployStorefront release 2026.05-demo shipped with a checkout regression.
- Failed CI/test runGitHub Actions posted a failed unit test run with two checkout failures.
- Latency spikeAPI checks degraded while checkout health checks started returning errors.
- Confirmed incidentErrorNotifier opened a critical incident after repeated external failures.
- Alert routedEmail, Slack, and a signed automation webhook received delivery records.
- ResolutionThe deploy was rolled back, recovery was confirmed, and the report budget was updated.
AI Intake
Agent-submitted error reports, improvement suggestions, and endpoint talk-back relayed from client sites and AI tools into this workspace for review.
Pending review0AI records do not change active systems until reviewed by humans or approved workflows.
Last 24h0
Bug reports0
Suggestions0
AI intake records are public-safe evidence. They can link to projects, domains, monitors, issues, incidents, tests, or releases, but the AI endpoint never approves repairs, closes issues, deletes records, rotates keys, deploys code, or claims certification.
No AI intake records yetClient-site agents can read /errornotifier-ai-endpoints.json on a WordPress site with the ErrorNotifier SDK and relay public-safe reports here.Open agent start guide