remove references without examples

Author: krisztianfekete

README.md

@@ -66,7 +66,7 @@ export OTEL_RESOURCE_ATTRIBUTES="agentevals.session_name=my-agent"
 python your_agent.py
 ```
 
-Traces stream to the UI in real-time. Works with LangChain, Strands, Google ADK, LlamaIndex, or any framework that emits OTel spans (`http/protobuf` and `http/json` supported). Sessions are auto-created and grouped by `agentevals.session_name`. Set `agentevals.eval_set_id` to associate traces with an eval set.
+Traces stream to the UI in real-time. Works with LangChain, Strands, Google ADK, or any framework that emits OTel spans (`http/protobuf` and `http/json` supported). Sessions are auto-created and grouped by `agentevals.session_name`. Set `agentevals.eval_set_id` to associate traces with an eval set.
 
 See [examples/zero-code-examples/](examples/zero-code-examples/) for working examples.
 

examples/README.md

@@ -21,7 +21,7 @@ The OTLP receiver runs on port 4318 (standard OTLP HTTP port) and accepts both `
 | [zero-code-examples/langchain/](./zero-code-examples/langchain/) | LangChain | OpenAI |
 | [zero-code-examples/strands/](./zero-code-examples/strands/) | Strands | OpenAI |
 
-This approach works with any framework that has OTel instrumentation: LangChain, Strands, Google ADK, LlamaIndex, Haystack, etc. If your framework already emits OTel spans, you only need to add `OTLPSpanExporter` (and `OTLPLogExporter` if it uses GenAI log-based content delivery).
+This approach works with any framework that has OTel instrumentation: LangChain, Strands, Google ADK, etc. If your framework already emits OTel spans, you only need to add `OTLPSpanExporter` (and `OTLPLogExporter` if it uses GenAI log-based content delivery).
 
 ### Resource attributes
 
@@ -80,7 +80,7 @@ Requires the `[streaming]` extra: `pip install "agentevals[streaming]"`. See [sd
 
 Trace format is **auto-detected**. Agents don't need to declare which format they use.
 
-- **OTel GenAI Semantic Conventions** (recommended for new agents). Standard `gen_ai.*` span attributes defined by the [OpenTelemetry GenAI working group](https://opentelemetry.io/docs/specs/semconv/gen-ai/). Framework-agnostic and interoperable. Works with LangChain, LlamaIndex, Haystack, Strands, and any framework that supports the conventions.
+- **OTel GenAI Semantic Conventions** (recommended for new agents). Standard `gen_ai.*` span attributes defined by the [OpenTelemetry GenAI working group](https://opentelemetry.io/docs/specs/semconv/gen-ai/). Framework-agnostic and interoperable. Works with LangChain, Strands, and any framework that supports the conventions.
 
 - **Framework-Native OTel Tracing**. Some frameworks (like Google ADK) emit their own proprietary span attributes. agentevals has dedicated converters for these formats.
 

examples/langchain_agent/README.md

@@ -10,7 +10,7 @@ A LangChain agent with:
 - Real-time trace streaming to agentevals dev server
 - Live conversation display and evaluation
 
-**Framework-Agnostic:** Uses standard OpenTelemetry GenAI semantic conventions that work with any framework (LangChain, LlamaIndex, Haystack, etc.).
+**Framework-Agnostic:** Uses standard OpenTelemetry GenAI semantic conventions that work with any framework (LangChain, Strands, etc.).
 
 ## Quick Start
 
@@ -225,7 +225,6 @@ response = llm.invoke(...)
 
 This pattern works with any framework using GenAI semantic conventions:
 
-- **LlamaIndex** - Enable OTEL, use same processors
-- **Haystack** - Enable OTEL instrumentation
+- **Strands** - Uses span events for message content
 - **Custom agents** - Use `opentelemetry-instrumentation-openai-v2`
 

src/agentevals/converter.py

@@ -2,7 +2,7 @@
 
 Supports two trace formats:
 1. ADK format (gcp.vertex.agent scope with ADK-specific attributes)
-2. GenAI semantic conventions (standard gen_ai.* attributes from LangChain, LlamaIndex, etc.)
+2. GenAI semantic conventions (standard gen_ai.* attributes from LangChain, Strands, etc.)
 
 Automatically detects the format and routes to the appropriate converter.
 Format detection checks span attributes and falls back to checking all spans if needed.

src/agentevals/genai_converter.py

@@ -2,8 +2,6 @@
 
 Supports traces from frameworks using OpenTelemetry GenAI semantic conventions:
 - LangChain (via LANGSMITH_OTEL_ENABLED)
-- LlamaIndex
-- Haystack
 - Any framework using standard gen_ai.* attributes
 """