slug: phoenix
provider: Arize Phoenix
generated_by: planning/capability-mapping/scripts/classify_capabilities.py
model: claude-opus-5
frame:
- Software & Technology
min_confidence: 0.7
capability_model:
  source: https://github.com/vincentmakes/turbo-ea-capabilities
  license: CC-BY-4.0
  attribution: Turbo EA Capabilities by Vincent Verdet — Turbo EA, https://github.com/vincentmakes/turbo-ea-capabilities,
    CC BY 4.0
  notice: NOTICE
edge_count: 5
edges:
- tag: Spans
  spec_file: phoenix-spans-api-openapi.yml
  capability_id: BC-4220.20
  capability_id_l1: BC-4220
  capability_name: Observability Management
  confidence: 0.8
  evidence: POST /v1/projects/{project_identifier}/spans  "Create spans"; GET /v1/projects/{project_identifier}/spans/otlpv1  "Search
    spans"; schemas OtlpEvent, OtlpKeyValue, SpanContext
  reason: Ingestion, search and annotation of OpenTelemetry spans — distributed tracing telemetry, which
    is squarely Observability Management (logs, metrics, traces).
- tag: Traces
  spec_file: phoenix-traces-api-openapi.yml
  capability_id: BC-4220.20
  capability_id_l1: BC-4220
  capability_name: Observability Management
  confidence: 0.8
  evidence: GET /v1/projects/{project_identifier}/traces  "List traces for a project"; schema TraceSpanData;
    DELETE /v1/traces/{trace_identifier}
  reason: Listing, annotating and deleting application traces built on OpenTelemetry — trace telemetry
    management, i.e. Observability Management.
- tag: Datasets
  spec_file: phoenix-datasets-api-openapi.yml
  capability_id: BC-610.60
  capability_id_l1: BC-610
  capability_name: Artificial Intelligence Management
  confidence: 0.72
  evidence: GET /v1/datasets/{id}/jsonl/openai_ft  "Download dataset examples as OpenAI fine-tuning JSONL
    file"; "Upload dataset from JSON, JSONL, CSV, or PyArrow"
  reason: Curation of example datasets used for LLM evaluation and fine-tuning — an AI/ML lifecycle (MLOps)
    artefact rather than enterprise master data or analytics reporting. Mapped to Artificial Intelligence
    Management with moderate confidence since 'dataset' could also read as generic data management.
- tag: Prompts
  spec_file: phoenix-prompts-api-openapi.yml
  capability_id: BC-610.60
  capability_id_l1: BC-610
  capability_name: Artificial Intelligence Management
  confidence: 0.72
  evidence: POST /v1/prompts  "Create a new prompt"; GET /v1/prompts/{prompt_identifier}/versions  "List
    prompt versions"; schemas PromptAnthropicOutputConfig, PromptGoogleInvocationParameters
  reason: Versioned prompt templates with per-LLM-provider invocation parameters — prompt asset lifecycle
    management within the AI/ML (LLMOps) practice. Not source control of product code, since the artefacts
    are model prompts and provider configs.
- tag: Experiments
  spec_file: phoenix-experiments-api-openapi.yml
  capability_id: BC-610.60
  capability_id_l1: BC-610
  capability_name: Artificial Intelligence Management
  confidence: 0.7
  evidence: POST /v1/datasets/{dataset_id}/experiments  "Create experiment on a dataset"; POST /v1/experiment_evaluations  "Create
    or update evaluation for an experiment run"
  reason: These are LLM/model evaluation experiment runs against datasets — model evaluation in the AI/ML
    lifecycle. Deliberately NOT product A/B testing (BC-4280.50), which concerns feature experiments on
    end users.