{"id":"dagster-api","name":"Dagster","homepage":"https://dagster.io","repo_url":"https://github.com/dagdagster-io/dagster","category":"data-pipeline","subcategories":["workflow-orchestration","data-engineering","asset-based"],"tags":["dagster","data-orchestration","asset-based","data-engineering","python","lineage"],"what_it_does":"Asset-based data orchestration platform that models pipelines as a graph of data assets with dependencies, offering a GraphQL API for pipeline management and observability in data platform engineering.","use_cases":["Triggering asset materialization runs programmatically for downstream pipeline automation","Querying asset lineage and materialization history for data observability workflows","Managing run schedules, sensors, and backfills via API","Building data platform dashboards from Dagster's rich pipeline metadata","Integrating data asset status into broader orchestration agents"],"not_for":["Non-Python data teams (Dagster is Python-centric)","Real-time stream processing (batch-oriented; use Kafka or Flink)","Simple scripts without asset dependency modeling needs","Organizations preferring simpler task-based orchestration over asset-centric model"],"best_when":"You're building a data platform with complex asset dependencies and want software engineering best practices (typing, testing, lineage) applied to data pipelines.","avoid_when":"Your data pipelines are simple, your team prefers task-based (not asset-based) thinking, or you already have Airflow or Prefect deeply integrated.","alternatives":["prefect-api","airbyte-api","dbt-api"],"af_score":75.5,"security_score":null,"reliability_score":null,"package_type":"mcp_server","discovery_source":["github"],"priority":"low","status":"evaluated","version_evaluated":"current","last_evaluated":"2026-03-01T09:50:05.458519+00:00","performance":{"latency_p50_ms":100,"latency_p99_ms":400,"uptime_sla_percent":99.9,"rate_limits":"No documented rate limits for self-hosted; Cloud has undisclosed limits","data_source":"llm_estimated","measured_on":null}}