raga-ai-hub/RagaAI-Catalyst

Health v2 · maintenance only

Adoption not part of the score

16150 stars · 3567 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

RagaAI Catalyst is a Python SDK for observability, monitoring, and evaluation of LLM and agentic AI applications, with tracing of agents, LLMs, and tool calls. It includes a self-hosted dashboard with timeline and execution graph views, plus dataset, prompt, guardrail, and red-teaming management.

Use cases

• trace and debug multi-agent LLM systems
• monitor LLM application performance in production
• evaluate LLM outputs against datasets
• manage and version prompts for LLM apps
• generate synthetic test data for LLM evaluation
• add guardrails and run red-teaming on AI agents
• visualize agent execution graphs and timelines

When to choose

• you are building LLM or multi-agent applications and need tracing and debugging
• you want a self-hosted observability dashboard for AI agents
• you need evaluation, guardrails, and red-teaming in one SDK

When to avoid

• you only need simple logging without LLM-specific analytics
• your stack is not Python-based
• you want a fully managed SaaS-only solution with no self-hosting

Facets

library · maturity active

monitoring tracing llm-inference agent-framework testing analytics data-visualization sdk large-language-models machine-learning developer-tools monitoring python self-hosted llm-observability agent-tracing llm-evaluation llmops guardrails prompt-management red-teaming synthetic-data ai-agents web-server

Member repositories

For agents

markdown · JSON · MCP: product_card(name="raga-ai-hub/RagaAI-Catalyst")