README.md

Integrations: "Works with" Guides

@icodenet/eval-dashboards is runner-agnostic and artifact-first.

If you can export rows and map them into eval-report/v1, you can use the same lint, check, report, history, and publish flow.

Supported import adapters in CLI today

eval-dashboards import --from=promptfoo --input=./promptfoo.json --out=.evals_output/promptfoo.json
eval-dashboards import --from=deepeval --input=./deepeval.json --out=.evals_output/deepeval.json
eval-dashboards import --from=openevals --input=./agentevals-like.json --out=.evals_output/openevals.json
eval-dashboards import --from=otel-genai --input=./otel-genai-spans.json --out=.evals_output/otel-genai.json

(openevals is an alias for the AgentEvals adapter.)

Works with pages

• Promptfoo
• DeepEval
• OpenAI eval surfaces / AgentEvals
• Anthropic eval methodology
• Langfuse
• W&B Weave
• Arize Phoenix
• Braintrust
• Ragas
• TruLens
• Patronus
• Trace/observability stacks
• OpenTelemetry GenAI evaluation events

Integration risk register

Before adding or changing an integration, check:

• Integration risk register