apoDemo · read-only
Sign in
Agent Testing
  • Tasks
  • Runs
  • Schedules
Observability
  • Traces
Tasks
  1. 1Open the failed runThe agent answered 'yes' — the benchmark's ground truth is 'Not Applicable'.
  2. 2Read the judgeJudged three times for stability — the Checks tab shows every pass and fail.
  3. 3Walk the traceThe Transcript tab holds the agent conversation behind the verdict.
  4. 4Compare modelsGLM vs DeepSeek on the same benchmark task — where they disagree is the story.
  5. 5See a human overrideThe judge failed this code review; a reviewer set it back to pass — then browse the real-agent tasks.

Captured example data · 2026-09-04

DocsGitHub
apo v0.2.0
Status
All tasks23 tasks
Pass
32%
dabstep-avg-fee-account-h-restaurantsFailed

Added 2026-09-17

Runs2
Avg time3m 3s
Avg cost$0.0379
Last run54d ago
Pass
50%
dabstep-avg-globalcard-fee-credit-10eurPassed

Added 2026-09-17

Runs3
Avg time8m 11s
Avg cost$0.0289
Last run54d ago
Pass
67%
dabstep-fee-delta-fee-384-jan-2023Errored

Added 2026-09-17

Runs2
Avg time8m 55s
Avg cost$0.0404
Last run54d ago
Pass
0%
dabstep-fee-ids-account-r-aci-bFailed

Added 2026-09-17

Runs2
Avg time44.4s
Avg cost$0.0113
Last run54d ago
Pass
50%
dabstep-fee-ids-belles-10-jan-2023Passed

Added 2026-09-17

Runs2
Avg time5m 51s
Avg cost$0.0795
Last run54d ago
Pass
50%
dabstep-fee-ids-belles-march-2023Failed

Added 2026-09-17

Runs2
Avg time7m 7s
Avg cost$0.0977
Last run54d ago
Pass
0%
dabstep-issuing-country-most-transactionsPassed

Added 2026-09-17

Runs2
Avg time15.9s
Avg cost$0.000909
Last run54d ago
Pass
100%
dabstep-lowest-fee-aci-januaryFailed

Added 2026-09-17

Runs3
Avg time9m 39s
Avg cost$0.0460
Last run54d ago
Pass
0%
dabstep-merchant-high-fraud-fine-riskFailed

Added 2026-09-17

Runs2
Avg time1m 39s
Avg cost$0.0162
Last run54d ago
Pass
0%
dabstep-top-ip-country-for-fraudFailed

Added 2026-09-17

Runs2
Avg time14.9s
Avg cost$0.000503
Last run54d ago
Pass
0%
Pass
17%
document-qaFailed

Added 2026-09-17

Runs3
Avg time26.8s
Avg cost$0.0354
Last run59d ago
real-agent/documents/document-qa
Pass
0%
data-extractionFailed

Added 2026-09-17

Runs3
Avg time27.6s
Avg cost$0.0462
Last run59d ago
real-agent/documents/data-extraction
Pass
33%
Pass
13%
api-testingFailed

Added 2026-09-17

Runs3
Avg time30.8s
Avg cost$0.0586
Last run59d ago
real-agent/engineering/api-testing
Pass
33%
bug-triageFailed

Added 2026-09-17

Runs3
Avg time32.0s
Avg cost$0.0524
Last run59d ago
real-agent/engineering/bug-triage
Pass
0%
code-reviewFailed

Added 2026-09-17

Runs3
Avg time47.6s
Avg cost$0.1169
Last run59d ago
real-agent/engineering/code-review
Pass
33%
config-generatorFailed

Added 2026-09-17

Runs3
Avg time31.3s
Avg cost$0.0509
Last run59d ago
real-agent/engineering/config-generator
Pass
0%
migration-plannerFailed

Added 2026-09-17

Runs3
Avg time51.8s
Avg cost$0.0992
Last run59d ago
real-agent/engineering/migration-planner
Pass
0%
count-dataset-tokens

Added 2026-09-17

Ready to runharbor/terminal-bench/count-dataset-tokens
data-extraction

Added 2026-09-17

Ready to runai-sdk-agent/data-extraction
data-extraction

Added 2026-09-17

Ready to runclaude-agent/data-extraction
Pass
0%
log-analyzerFailed

Added 2026-09-17

Runs3
Avg time38.8s
Avg cost$0.1036
Last run59d ago
real-agent/operations/log-analyzer
Pass
0%
Pass
0%
research-synthesisFailed

Added 2026-09-17

Runs3
Avg time45.2s
Avg cost$0.0649
Last run59d ago
real-agent/research/research-synthesis
Pass
0%
Pass
0%
security-auditFailed

Added 2026-09-17

Runs3
Avg time35.8s
Avg cost$0.0613
Last run59d ago
real-agent/security/security-audit
Pass
0%