trueraOfficial·8 Agent Skills Included

trulens

Trace, evaluate, and improve LLM apps and AI agents

Instruments LLM apps and agents with OpenTelemetry tracing to capture every step, then scores outputs with LLM judges for relevance, groundedness, and tool use. Eliminates guesswork when debugging hallucinations, bad retrievals, and failing agent tool calls. Compares app versions on cost, latency, and quality so teams ship the best-performing configuration.

3,531|335

↗

npx skills add truera/trulens --all -g -y

Available:

Gives the AI agent the project structure, instrumentation patterns, and evaluation code conventions it needs to correctly set up TruLens tracing and feedback functions in user apps.

All Skills in This Repository (8)

Pure Emerald Level Indicators

📦 In Repo

truera

trulens-blocking-guardrails

Configure TruLens feedback functions as runtime guardrails to block unsafe inputs, outputs, and irrelevant context.

trulens-diagnosis

Diagnose low LLM evaluation scores and generate targeted improvement recommendations.

trulens-evaluation-workflow

Orchestrates end-to-end TruLens evaluation workflows for LLM apps.

trulens-dataset-curation

Create and curate ground-truth evaluation datasets for TruLens.

trulens-instrumentation

Instrument LLM apps to capture traces for evaluation and debugging.

trulens-running-evaluations

Orchestrate TruLens evaluations across apps and versions with TruChain, TruGraph, and TruLlama.

trulens-evaluation-setup

Configure TruLens feedback functions and selectors for LLM output evaluation.

trulens-notebook-execution

Execute and display TruLens notebooks using jupyter nbconvert.

Official

Intermediate

Frequently Asked Questions

FAQPage Schema

How to install TruLens?▼

Run `npx skills add truera/trulens --all -g -y` in your terminal to install all TruLens skills globally for your AI agent.

How to evaluate a RAG app with AI?▼

TruLens wraps your LangChain or LlamaIndex app, traces each retrieval and generation step, and scores the RAG triad: context relevance, groundedness, and answer relevance.

How to debug why my AI agent fails?▼

TruLens records every tool call, plan step, and LLM generation as a trace, then applies agent metrics like Tool Selection and Plan Adherence to pinpoint the exact failing step.

Does TruLens work with LangChain and LlamaIndex?▼

Yes. It provides ready-made wrappers like TruChain, TruGraph, and TruLlama that auto-instrument apps built with LangChain, LangGraph, and LlamaIndex.

Can TruLens block harmful outputs in real time?▼

Yes. Its guardrails skill turns feedback functions into runtime checks that block toxic inputs, filter irrelevant retrieved context, and stop unsafe outputs before users see them.