apo
Demo · read-only
Sign in
Agent Testing
Tasks
Runs
Schedules
Observability
Traces
Toggle Sidebar
demo
Runs
Runs
demo-bch
real-agent/documents/data-extraction
demo-run_2
Failed
real-agent/documents/data-extraction
·
anthropic/claude-sonnet-4.5
·
cli
·
8w ago
Details
▸
Task
Run
Trace home
Delete
80%
pass rate
4
passed
·
1
failed
·
5
checks
43.3s
duration
$0.1106
25.7k tok
—
reasoning
Checks
5
Conversation History
Deliverables
Trace home
4
/5 passed
1 failed
Click to expand
✓
used-extraction-workflow
✓
extracted-all-key-fields
✗
findings-grounded-in-invoice
The evaluation requires checking if every number, date, name, and email in the findings can be traced to the invoice document content. However, no invoice document content has been provided for comparison. The values presented include:
✓
extraction-is-thorough
✓
invoice-file-present