[bot] Add instrumentation support for langchain-rust LLM framework

[braintrust-bot]

Summary

The Braintrust Rust SDK has no instrumentation support for langchain-rust, the most widely adopted LangChain implementation for Rust. Braintrust already wraps LangChain in four other language SDKs — Python (langchain-core), TypeScript (callback handler), Go (langchaingo), and Java (langchain4j) — but the Rust SDK has no equivalent.

What is missing

The langchain-rust crate provides execution-oriented API surfaces that could be instrumented to automatically capture spans:

• Chains — LLMChain, ConversationalChain, SequentialChain, and other chain types that compose LLM calls
• Agents — agent execution with tool calling and multi-step reasoning
• LLM calls — direct model invocations across providers (OpenAI, Anthropic Claude, Azure OpenAI, Ollama, MistralAI)
• Embeddings — embedding generation across providers (OpenAI, Azure OpenAI, Ollama, FastEmbed, MistralAI)
• Retrievers — retrieval operations for RAG workflows

Braintrust's other SDKs instrument LangChain via callback handlers that capture chains, agents, retrievers, tools, and individual LLM calls as traced spans.

Library significance

• 1.3k GitHub stars, 532 commits on main branch
• 123k+ total downloads on crates.io
• Active community with 55 open issues and 28 pull requests
• Latest version: 4.6.0
• Supports multiple LLM providers (OpenAI, Anthropic, Azure, Ollama, MistralAI) and vector stores (OpenSearch, Postgres, Qdrant, SQLite, SurrealDB)
• Official docs: https://docs.rs/langchain-rust

Braintrust docs status

supported (in other languages) — Braintrust documents LangChain integration for Python (langchain-core ≥ 0.3.28), Go (langchaingo ≥ 0.1.13), Java (langchain4j ≥ 1.8.0), and TypeScript (callback handler). Rust is not listed as a supported language for LangChain tracing.

• https://www.braintrust.dev/docs/instrument/trace-llm-calls — lists LangChain support for Python, Go, Java
• https://www.braintrust.dev/docs/integrations/sdk-integrations/langchain — LangChain integration guide covering TS, Python, Go

Upstream sources

• GitHub repository: https://github.com/Abraxas-365/langchain-rust
• API documentation: https://docs.rs/langchain-rust/latest/langchain_rust/
• Crate page: https://crates.io/crates/langchain-rust

Local files inspected

• src/extractors.rs — only OpenAI and Anthropic usage extractors; no LangChain-related code
• src/stream.rs — stream aggregator for OpenAI Chat Completions format only; no callback handler
• src/lib.rs — public API exports; no LangChain references
• Cargo.toml — no langchain-rust dependency
• Full codebase grep for "langchain", "chain", "agent", "callback" — zero results related to LangChain integration