{"categories":[{"label":"Monitoring","url":"https://skillfed.io/packages/category/system-monitoring/3"}],"enrichment":{"capability":"Auto-instruments Haystack LLM pipelines and components to emit OpenTelemetry traces for monitoring and observability in tools like Arize Phoenix.","skillfed_tags":["llm-observability","opentelemetry","haystack"],"use_cases":["Monitor Haystack LLM pipeline execution in real time by sending traces to Arize Phoenix running locally on your development machine.","Debug slow or failing Haystack components by inspecting detailed span attributes (prompts, model calls, latencies) in an observability UI.","Track production Haystack applications by exporting traces to a centralized OpenTelemetry collector for long-term analysis and alerting.","Correlate Haystack pipeline behavior with upstream or downstream systems by using OpenTelemetry's trace context propagation.","Audit LLM application behavior by capturing what prompts were sent, which models were called, and how long each step took."],"what_it_does":"This package wraps Haystack pipelines and components with OpenTelemetry instrumentation, automatically capturing execution traces and sending them to an observability backend. It integrates with the OpenInference semantic conventions to standardize how LLM application traces are structured, making it easy to monitor Haystack-based LLM systems in tools like Arize Phoenix or Arize AX without modifying your application code.\n\nThe instrumentation is applied once via a single call to HaystackInstrumentor().instrument(), after which all Haystack components in your pipeline emit traces automatically. You configure where traces are sent by setting up an OpenTelemetry tracer provider with a span processor and exporter (typically OTLP HTTP to a local or remote collector). The package depends on openinference-instrumentation, opentelemetry-api, opentelemetry-instrumentation, and their semantic convention libraries to handle the tracing plumbing.","worth_installing":"Yes. The package is actively maintained (release 7 days ago), production-stable, has no known vulnerabilities, and low install friction. It solves a real observability gap for Haystack users. Install it if you need to monitor or debug Haystack LLM pipelines; skip it only if you have no observability requirements or use a different instrumentation approach."},"id":"openinference-instrumentation-haystack","links":{"html":"https://skillfed.io/packages/openinference-instrumentation-haystack","md":"https://skillfed.io/packages/openinference-instrumentation-haystack.md","pypi":"https://pypi.org/project/openinference-instrumentation-haystack/"},"maintenance":{"status":"active"},"meta":{"latest_release":"2026-08-07","license_spdx":"Apache-2.0","license_treatment":"permissive","name":"openinference-instrumentation-haystack","python_support":"supports_current","summary":"OpenInference Haystack Instrumentation"},"popularity":{"monthly_downloads":333652,"position":7498,"tier":"top_15000"},"security":{"n_vulnerabilities":0},"version":"0.1.36"}