MCP For Claude Marketplace
LangSmith

LangSmith MCP Connector for Claude

A+

Observability and evaluation platform for LLM applications — monitor traces, debug agent runs, and track performance metrics across your AI stack.

3 tools Official Updated Oct 1, 2026 Official Vinkius Partner
More Details Connect to Claude

Connect your AI agent to LangSmith — the observability platform from the LangChain team that gives you complete visibility into your LLM applications.

What you can do

  • List Projects — View all tracing projects with aggregate metrics: total runs, median latency, feedback scores, and creation dates
  • List Runs — Browse recent traces in any project. See run names, types (LLM, chain, tool), status (success/error), token usage, and timing
  • Run Details — Deep-dive into any specific run to see its full execution trace, inputs, outputs, and associated feedback

How it works

  1. Subscribe to this server
  2. Enter your LangSmith API key (5,000 free traces/month)
  3. Your agent can now monitor and debug LLM applications

Who is this for?

  • AI Engineers — monitor LLM calls, chains, and agent actions in production
  • ML Teams — track experiment performance, compare model outputs, and identify regressions
  • DevOps — set up alerts for error rates, latency spikes, and cost anomalies in AI workloads
llm-observabilitytracingevaluationperformance-metricsai-debuggingprompt-testing

3 tools expose this connector's capabilities to your AI agent.

langsmith_list_projects

Each project groups related traces together and shows aggregate metrics like total runs, median latency, and feedback counts. List all tracing projects in your LangSmith account with run counts, latency stats, and feedback metrics

langsmith_get_run

Useful for debugging specific LLM calls or agent actions. Get detailed information about a specific run/trace by its ID

langsmith_list_runs

Each run represents a single LLM call, chain execution, or agent action. Shows status (success/error), latency, and token consumption. List recent traces/runs in a specific LangSmith project. Shows run names, types, status, token usage, and timing

See how to talk to your AI agent using LangSmith.

List all my LangSmith projects and show their metrics.

Found 4 projects: | Name | ID | Runs | Latency | Created | |---|---|---|---|---| | production-agent | `a1b2c3d4` | 12,450 | 340ms | 2026-01-15 | | staging-chatbot | `e5f6g7h8` | 3,200 | 280ms | 2026-02-20 | | research-rag | `i9j0k1l2` | 890 | 520ms | 2026-03-01 | | test-evaluations | `m3n4o5p6` | 150 | 190ms | 2026-03-28 |

Show me the last 5 runs in my production-agent project.

Last 5 runs in 'production-agent': | Name | ID | Type | Latency | Created | |---|---|---|---|---| | agent_executor (success) | `r1s2t3` | chain | 1,250 tokens | 2026-04-04 | | gpt-4o (success) | `u4v5w6` | llm | 890 tokens | 2026-04-04 | | tool:web_search (success) | `x7y8z9` | tool | 340ms | 2026-04-04 | | agent_executor (error) | `a0b1c2` | chain | 450 tokens | 2026-04-03 | | gpt-4o (success) | `d3e4f5` | llm | 1,100 tokens | 2026-04-03 |

Get details on the failed run a0b1c2.

Run details for `a0b1c2`: | Name | ID | Type | Latency | Created | |---|---|---|---|---| | agent_executor \| chain \| error | `a0b1c2` | chain | 450 tokens | 2026-04-03 | Error: Tool 'web_search' returned timeout after 30s. The agent retried 3 times before failing.

LangSmith is the 'Datadog for LLM applications'. Without observability, AI agents in production are black boxes — you can't see what they're doing, why they fail, or how much they cost. LangSmith traces every LLM call, chain execution, and tool use, giving you complete visibility into inputs, outputs, latency, token usage, and error rates.

Related Connectors

Lancerkit

Lancerkit MCP

10 tools

Connect Lancerkit to any AI agent via MCP.

A+ View details →
Dagster

Dagster MCP

6 tools Official

Orchestrate data pipelines via Dagster — monitor jobs, track runs, manage software-defined assets, and audit schedules directly from any AI agent.

A+ View details →
Storylane

Storylane MCP

12 tools Official

Interactive, personalized product demos.

A+ View details →
Helicone (LLM Observability)

Helicone (LLM Observability) MCP

10 tools Official

Monitor LLM usage via Helicone — track requests, analyze costs, measure latency, and manage prompts.

A+ View details →