The Stack · Observability & evals (managed) API

LangSmith

LangChain's managed tracing, evaluation and prompt-engineering platform for LLM and agent apps.

API keyFreemium🔵 agent: self serve instant keyMCP ✓

Get API key → instant key, usually no card Docs Website

CategoryObservability & evals (managed)

TypeAPI service

Authapi-key

PricingFreemium

Agent signup🔵 self serve instant key

SDKsPython, TypeScript, REST

MCP serveravailable →

Pricing: Free Developer: 5k traces/mo, 14-day retention, 1 seat; Plus $39/seat/mo (10k traces incl., $2.50/1k overage); Enterprise custom.

🔵 Agents: self serve instant key

Instant self-serve — free signup gives a key immediately (a human unblocks an agent in under 2 minutes).

Free signup via Google/GitHub/email, no credit card; create an API key in Settings instantly. No public account-provisioning API for autonomous agents.

LangSmith — LangChain's managed tracing, evaluation and prompt-engineering platform for LLM and agent apps.
Website: https://www.langchain.com/langsmith
Docs: https://docs.langchain.com/langsmith/observability
Get a key: https://smith.langchain.com/
Agent signup: self serve instant key — Free signup via Google/GitHub/email, no credit card; create an API key in Settings instantly. No public account-provisioning API for autonomous agents.
Auth: api-key
MCP server: https://github.com/langchain-ai/langsmith-mcp-server
Pricing: freemium — Free Developer: 5k traces/mo, 14-day retention, 1 seat; Plus $39/seat/mo (10k traces incl., $2.50/1k overage); Enterprise custom.
Full record: https://dreaming.press/api/tools/langsmith.json

Machine record: /api/tools/langsmith.json · add its MCP server →

Quickstart

import openai
from langsmith import wrappers, traceable

# set LANGSMITH_TRACING=true and LANGSMITH_API_KEY=lsv2_...
client = wrappers.wrap_openai(openai.OpenAI())

@traceable
def answer(question: str):
    return client.chat.completions.create(
        model="gpt-4o-mini",
        messages=[{"role": "user", "content": question}],
    )

answer("What is LLM observability?")

python · full docs →

Model Context Protocol

LangSmith exposes an MCP server, so you can add its tools to Claude, Cursor, or any MCP client: https://github.com/langchain-ai/langsmith-mcp-server

LangSmith in our coverage

LangGraph Checkpointer: Postgres vs Redis Backend Comparison

2026-07-08

Tracing MCP Tool Calls Without Sessions: Why traceparent Became the Correlation ID

2026-07-08

OpenTelemetry for AI Agents: The Span Tree Is Stable, the Attributes Aren't

2026-07-07

How to Version Prompts in Production AI Agents: A Prompt Change Is a Deploy

2026-07-05

LangGraph Platform Is Now LangSmith Deployment — and Your Agent Ships as an MCP Server by Default

2026-07-05

TensorZero Shut Down With Money in the Bank: What the LLMOps Squeeze Looks Like

2026-07-04

LangSmith FAQ

What is LangSmith?

LangSmith is LangChain's managed tracing, evaluation and prompt-engineering platform for LLM and agent apps. It's in the Observability & evals (managed) category of the dreaming.press tool directory.

Is LangSmith free?

LangSmith is freemium — free to start, paid plans for scale. Free Developer: 5k traces/mo, 14-day retention, 1 seat; Plus $39/seat/mo (10k traces incl., $2.50/1k overage); Enterprise custom..

Can an AI agent sign up for LangSmith automatically?

Almost — signup is instant and self-serve (a free key with no sales call), so a human can unblock an agent in under two minutes. Free signup via Google/GitHub/email, no credit card; create an API key in Settings instantly. No public account-provisioning API for autonomous agents..

Does LangSmith have an MCP server?

Yes — LangSmith offers a Model Context Protocol server, so you can add its tools to Claude, Cursor, or any MCP client: https://github.com/langchain-ai/langsmith-mcp-server

Is LangSmith open source?

LangSmith is a hosted API service, not an open-source project.

What languages does LangSmith support?

LangSmith offers Python, TypeScript, REST.