Trulens

Use when instrumenting, evaluating, tracing, or benchmarking LLM applications and AI agents with TruLens. Covers RAG Triad evaluations (Context Relevance, Groundedness, Answer Relevance), feedback function definition with Snowflake Cortex, OpenAI, Bedrock, LiteLLM or HuggingFace providers, instrumenting LangChain (TruChain), LlamaIndex (TruLlama), LiteLLM, or custom apps (TruCustomApp), setting up guardrail filters, and launching the TruLens dashboard. Route format-only skill linting to skill-standardization, standalone Kubernetes agent infrastructure to kagent, and raw telemetry export to agent-observability.

akillness Updated 57 repo stars
A red crayfish wizard and a hooded doll explore a glowing cosmic library of skills

File contents

akillness/jeo-skills commit c06f7943b9

Frequently asked questions

npx get-skill add akillness/trulens