Skill UI

Deliver production-ready LLM applications, RAG systems, and multi-agent orchestration with vector databases, embedding strategies, safety controls, and observability for enterprise deployments.

Deliver production-ready LLM applications, RAG systems, and multi-agent orchestration with vector databases, embedding strategies, safety controls, and observability for enterprise deployments.

Migration monitoring, CDC, and observability infrastructure

Provides Databricks monitoring instructions using system tables, SQL alerts, and the SDK to track job health, cluster utilization, query performance, costs, and exportable metrics for observability-driven operations.

Automate deploying production-ready monitoring stacks like Prometheus, Grafana, and Datadog, delivering configuration files, deployment scripts, and guided instructions for infrastructure-focused observability setups.

Provides observability implementation guidance for Guidewire InsuranceSuite, covering structured logging, metrics collection, distributed tracing, and alerting to monitor deployments via Datadog/Splunk/New Relic.

Manage Langfuse trace exports, retention policies, GDPR compliance, anonymization, and audit logging to keep observability data secure while fulfilling organizational and regulatory requirements.

Guided process for executing complex Langfuse migrations, covering data export/import, dual-write zero-downtime, validation, and rollbacks, ideal when moving between observability platforms or Langfuse environments without losing trace continuity.

Guides teams through Langfuse security best practices, covering API key handling, PII scrubbing before tracing, access control on self-hosted deployments, and retention cleanup to keep LLM observability data private.

This skill acts as an SLO (Service Level Objective) expert, specializing in implementing reliability standards and error budget-based engineering practices. It guides the design of comprehensive SLO frameworks, establishment of meaningful SLIs, and creation of monitoring systems. Use this to balance product feature velocity with required service reliability, ensuring all engineering decisions are data-driven and aligned with business objectives.

Phoenix is an open-source observability platform for tracing, evaluating, and monitoring LLM applications; use it to debug prompts, run dataset evaluations, build experiment pipelines, and watch production inference in real time.

A comprehensive suite of 25 advanced agent skills designed for modern software development and platform operations. These tools cover the entire development lifecycle, including multi-agent architecture design, robust CI/CD pipelines, database schema management, performance profiling, observability setup, and security auditing. Ideal for complex system integration and platform engineering teams.