Projects/Customer Support Reply

Customer Support Reply

Active

First-response generator with KB retrieval

Open reports

Project Settings

Identity

Name*?Human-readable project name shown in lists and reports.

Description?One-line summary of what AI behavior this project evaluates.

Owner*?Team or person responsible for this project's evaluations.

Status?active: running evaluations. paused: no new runs. archived: read-only.

Evaluation Defaults

Default model*?The AI model being evaluated. Used as default for new runs.

Active rubric?Rubric used for scoring runs. Changing this affects all future runs — existing run scores are unaffected.

Judge model?LLM used as evaluator for llm_judge dimensions. Leave empty to use the same model as default.

Metadata

Tags?Comma-separated labels for filtering and grouping projects.

Notes?Free-text field for team context. Not shown in public reports.

Project Intelligence

Computed from stored runs, cases, rubrics, and safety findings.

Evaluation Coverage

Total cases1

Passing0

Failing1

Unscored0

Regression cases0

Safety cases1

Recent Quality

Total runs1

Last run score0.64

Previous score—

Score delta—

Regression statusClean

Human Review

Open items1

P0 safety0

P1 low-confidence0

Reviewed cases1

Safety State

Open blockers1

PII findings0

False confirmations0

Policy findings1

Active Rubric

NameCustomer Support Reply

Versionv1.0

Dimensions8

Weights normalizedNo

Safety gateEnabled

Min threshold≥ 0.70