OpenTelemetry plugins

Build, deploy, and monitor AI-powered cloud applications on Azure using containerized apps, serverless functions, OpenAI integration, AI Search, and observability across .NET, Python, and Node.js.

Diagnose performance bottlenecks, implement distributed tracing, and manage incident response with Prometheus, Grafana, OpenTelemetry, and Datadog. Define SLIs/SLOs, run blameless postmortems, and build production-ready observability pipelines for microservices and infrastructure.

Orchestrate multi-agent teams for complex AI-driven projects: decompose tasks, match capabilities, coordinate workflows, manage shared context and errors, distribute workloads, monitor performance with Prometheus and OpenTelemetry, and synthesize insights from interactions. Integrates PowerShell, .NET, Azure ops via specialist subagents.

Evaluate and improve LLM applications by instrumenting agents, chatbots, and RAG pipelines with DeepEval tracing, generating test suites, running evaluations, and exporting traces to Confident AI for observability and iterative refinement.

Set up distributed tracing for microservices using OpenTelemetry with Jaeger or Zipkin backends. Automate SDK integration, service instrumentation, context propagation, span creation, trace sampling, collection, and dashboard deployment for end-to-end request visibility and performance analysis.

Integrate Azure services into Java apps using SDK clients and best practices for AI agents, anomaly detection, vision analysis, document processing, real-time voice/chat/SMS, storage/blobs/tables/Cosmos DB, Event Hubs/Grid, Key Vault secrets/keys, authentication, monitoring, and batch compute with sync/async operations.

Use gcx CLI to debug Grafana observability stacks: investigate alerts, SLO breaches, synthetic check failures via Prometheus metrics and Loki logs; manage dashboards, SLOs, resources with GitOps; scaffold Go projects; automate setups and code generation for resources-as-code.

Set up SAP BTP Cloud Logging instances via BTP Cockpit, CF CLI, BTP CLI, or Service Operator. Configure log ingestion from Cloud Foundry, Kyma runtimes, OpenTelemetry, or JSON API. Analyze logs, metrics, and traces in OpenSearch Dashboards. Secure with SAML authentication, manage certificates, and troubleshoot ingestion issues.

Instrument Python and JavaScript LLM apps with LangSmith tracing using LangChain auto-tracing, decorators, or OpenTelemetry; create, manage, and upload evaluation datasets; build custom evaluators like LLM-as-Judge; run evaluations locally via SDK or CLI, and query/export traces.

Orchestrate Station AI agents: create, run, update agents; manage environments, MCP servers, workflows; execute tasks with full access to 55+ MCP tools; debug operations; and export traces, metrics, logs via OpenTelemetry to observability backends like Honeycomb or DataDog.

Set up local-first LLM agent observability with TMA1 and GreptimeDB, including OTLP telemetry export from Claude Code and webhook configuration. Query costs, token usage, traces, events, errors, model comparisons, and tool activity via SQL on GreptimeDB using curl.

Interact with Elasticsearch and Kibana via curl REST API to query using Query DSL or ES|QL, index and manage documents with CRUD operations, configure mappings and ILM policies, run aggregations, monitor cluster health, deploy dashboards, integrate OpenTelemetry patterns, and troubleshoot issues.

Adopt OpenTelemetry observability across your stack: configure and deploy the Collector, instrument applications in multiple languages, write and debug OTTL transformations, and validate attribute conventions.

Delegate observability implementation to expert agents that handle OpenTelemetry instrumentation for distributed tracing, structured logging pipelines with tools like Vector and Loki, Prometheus metrics and alerting, Grafana dashboards, SLO definitions, and incident response workflows for optimized system debugging.

Manage the full Arize AX observability lifecycle for LLM applications: auto-instrument code, export traces, create and evaluate datasets, run experiments, optimize prompts, and audit compliance — all through the ax CLI.

Autonomously design, configure, deploy, and troubleshoot production-grade AI agents on AWS using Bedrock, AgentCore, and Strands Agents SDK with Terraform-first infrastructure and CloudWatch/OpenTelemetry observability, every recommendation traced to official AWS sources.

Enforce custom protocols on Claude Code file operations before Edit, Write, or NotebookEdit tools run, capture OTLP traces after every tool use for OpenTelemetry export, and track sessions to monitor AI-assisted coding workflows.

Delegate SDLC workflows to specialist AI agents that architect cloud-native systems, design databases, conduct deep web research, optimize performance and observability, distill repo knowledge, and build production agents via orchestrated pipelines.

Query local Lensflare telemetry datasets directly in Claude Code to analyze traces span-by-span for debugging slow requests and errors, summarize recent failures with root causes across services, and translate natural language questions into executed log and metric queries with summarized insights.

Helps instrument and configure OpenTelemetry telemetry pipelines across languages and the Collector: SDK setup, YAML config, OTTL, version compatibility, semantic conventions, migration patterns, and synthetic data generation.

Analyze AI agent execution traces in OTEL JSON or Claude Code JSONL format to detect issues like goal drift, grounding failures, missed actions, guardrail violations, and instruction following errors. Triage findings with specialized agents, generate and review reports, remediate context via diffs to prompts and tools, and enable autosync for ongoing monitoring from LangSmith or LangFuse sources.

Review Spring Boot codebases using skill detection and parallel agents for DDD patterns, security, testing, observability, and Modulith modules. Verify upgrades to Spring Boot 4 with multi-phase analysis producing severity-categorized migration checklists and remediation steps. Implement data layers, REST APIs, and configurations via guided skills.

Auto-instrument Python, JavaScript/TypeScript, and Rust apps with Logfire observability for traces, logs, and metrics using OpenTelemetry SDKs. Debug production errors by querying traces, running SQL on OTEL data, suggesting fixes, and sharing links. Start local dev sessions to provision credentials and send traces to Logfire UI. Connect remotely via MCP for logs, traces, and spans.

Develop, review, and deploy Go projects with conventions for architecture, testing, and git workflow, while also building interactive web UIs with Datastar, performing security reviews, fine-tuning AI models, and maintaining living documentation and experimental optimization loops.

Enforce expert-level programming principles during refactoring, testing, API design, code review, and commits using rigid checklists distilled from '97 Things Every Programmer Should Know'.

Investigate observability stacks by querying traces, logs, and metrics in OpenSearch with PPL and Prometheus with PromQL, correlating via OTel conventions from metric spikes to error logs, checking component health, and defining SLOs/SLIs.

Enforce a complete TDD workflow in Claude Code: detect bugs and test gaps, refactor code smells, eliminate duplication, optimize performance, and review security threats — all with multi-cycle auto-fixing and parallel agent execution

Load before Deno development to access knowledge of versions 2.2-2.7 features like permission sets, deno audit, OpenTelemetry integration, Deno.spawn(), QUIC support, test hooks, and dx command, enabling work on modern Deno projects without knowledge gaps.

Manage the full lifecycle of TrueFoundry AI Gateway — configure model routing, guardrails, MCP servers, prompt registry, and Skills Registry; migrate existing codebase to gateway; diagnose issues with logs and traces.

Develop Effect-TS applications using guided skills for effects, schemas, streams, layers, concurrency, error handling, and testing; run automated compliance checks with fixes; migrate code from Promises, async/await, or fp-ts; perform strict reviews and parallel git tasks with specialized agents.

Deploy and manage OpenTelemetry Collector pipelines shipping to Coralogix, instrument applications with OTel SDKs, write and debug OTTL transformations, and resolve telemetry semantic issues across Kubernetes and cloud environments.

Apply OpenTelemetry conventions to name spans, set semantic attributes and status codes, define instrumentation boundaries, and test traces in-memory without external collectors. Use for setting up distributed tracing or validating trace instrumentation in code.

Configure OpenTelemetry telemetry for Claude Code in Claudicle by writing environment variables and resource attributes to .claude/settings.local.json using a single CLI command. Supports --disable to turn off telemetry.

Run production incident investigations and SLO monitoring in Honeycomb directly from Claude Code — query traces and metrics, analyze root causes with BubbleUp, interpret SLO burn rates, and instrument applications with OpenTelemetry.