Skill Detail

Evaluate and trace agent workflows with RagaAI Catalyst

Use RagaAI Catalyst to instrument agent runs, collect traces, manage evaluation datasets, inspect execution graphs, and review regressions before deployment.

Monitoring & AlertsMulti-Framework
Monitoring & Alerts Multi-Framework Security Reviewed
⭐ 16.2k GitHub stars
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill evaluate-and-trace-agent-workflows-with-ragaai-catalyst Copy
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Tools required
Python environment, ragaai-catalyst SDK, Catalyst project credentials, instrumented LLM or agent workflow, and evaluation datasets or trace samples.
Install & setup
Install the ragaai-catalyst Python package, configure a Catalyst project, instrument agent or LLM calls with the SDK, then run representative workflows and review traces and evaluations in Catalyst.
Author
raga-ai-hub
Publisher
Open Source
Last updated
Jun 11, 2026
Quick brief

Invoke this workflow when a deployed or pre-release agent workflow needs evidence about what happened inside a run. The operator creates a Catalyst project, instruments LLM, tool, and agent steps with the Python SDK, attaches datasets or evaluation cases, and reviews traces, timelines, metrics, guardrails, and execution graphs. Reviewable inputs include the project configuration, instrumented agent code, evaluation dataset, prompt or tool versions, and the run set under review. Reviewable outputs include trace records, evaluation results, failure clusters, timeline views, and monitoring signals that can be used for release decisions.

How it works

What this skill actually does

Use this instead of manually reading scattered logs when debugging multi-step agents, comparing prompt changes, checking guardrail behavior, or benchmarking tool-use quality. The scope boundary is observability and evaluation for agent and LLM workflows. It is not a generic analytics platform card; the skill is specifically about using Catalyst’s documented SDK and dashboard workflow to trace, evaluate, and review agent behavior with reproducible evidence.