LangSmith MCP server: 15 tools, now archived

LangSmith exposes data about your own agents, and its repository is archived, so the package will not change.

Last verified 1 September 2026 · from LangChain's repository, PyPI and the server's own tools/list response

This page is one of 90 in a directory of monitoring and developer tools. Each page checks what an AI coding agent can get out of the tool through its MCP server, the connector that lets the agent query the tool directly.

What is the LangSmith MCP server?

It is LangChain's server for the Model Context Protocol. LangSmith records what your own LLM applications did, so the agent gains tools for reading those runs back.

The surface splits into five areas: prompts, traces and runs, datasets and examples, experiments, and billing usage. That last one is unusual among agent-observability connectors, which mostly stop at telemetry.

It runs locally and authenticates with a LangSmith API key. There is no hosted endpoint here; the hosted product is LangSmith itself, and this is a client for it.

LangSmith covers agent observability broadly, reaches part of the way into release through prompt management and into cost through billing usage, and reaches nothing in code review or production observability.

For an agent, LangSmith answers questions about how your LLM application behaved and what you fed it, and nothing about the machines underneath. An agent connected here can pull the runs behind a bad answer and the dataset a prompt was tested against. It cannot tell you the database was slow, so this connector pairs with an infrastructure one.

What can an AI agent do with LangSmith?

What connecting costs before the first question

A context window is the amount of text a model can hold at once, and connecting uses some of it before the agent asks anything. Every tool the server registers sends its name, description and input schema to the model when the session starts. The tool list is charged whether or not a single tool is called.

LangSmith is verbose for its size: 15 tools cost about 5,458 tokens. Arize Phoenix registers 27 tools and costs less. LangSmith's tool descriptions average 1,120 characters. 1,120 characters is long for a tool description.

Encodings: GPT-5.6, GPT-5.5 o200k_base; GPT-4, GPT-3.5 cl100k_base; Claude Haiku 4.5 to Opus 5 count_tokens. Counted 21 August 2026 from tools/list on langsmith-mcp-server 0.1.1. It was run locally with a placeholder key.

Registration is static, so no LangSmith account was involved. Figures cover name, description and input schema.

The two models in column one share the o200k_base encoding, so they get one count. The spread is formatting, about 22%. Reasoning tokens are output, spent per request. The tool list is input, charged once at connection.

The Claude figures are measured rather than encoded. Anthropic publishes no local tokenizer. So the definitions were sent to the API's count_tokens endpoint, and a no-tools baseline was subtracted. The range covers two tokenizer generations rather than two serializations. Haiku 4.5 and Opus 4.6 sit at the low end. Opus 5 and Sonnet 5 sit at the high end. The gap between them is about a quarter.

Can LangSmith watch your own AI agents?

LangSmith is built for tracing agents, so this table asks which of those signals the connector exposes. Counted from the server's own tools/list response on 21 August 2026.

LangSmith MCP server limits: where answers come back incomplete

The first limit below is about the repository rather than the connector, which still works exactly as documented.

How to configure LangSmith MCP for agents

1. 1

   Pin the version

   The repository is archived and the package sits at 0.1.1, so what it exposes is what it will expose. Pinning avoids a surprise if the package is ever yanked.

2. 2

   No allowlist is needed for safety

   Every tool reads. The four documentation-only tools are worth disabling for context rather than for risk: they return long markdown pages an agent rarely needs.

3. 3

   Set the workspace

   LANGSMITH_WORKSPACE_ID scopes what the agent sees; without it a multi-workspace account can return results from the wrong place.

4. 4

   Point at the right host

   LANGSMITH_ENDPOINT is needed for self-hosted or non-default LangSmith instances.

Paste this into your agent’s instructions

LangSmith's quiet failure is scope: a run search in the wrong project, or a thread with no identifier, both return nothing. The block makes the agent name what it searched.

When you answer using LangSmith MCP tools, state:
- The project and workspace you searched.
- The time range, and how many runs you actually retrieved.
- Whether a thread identifier existed. get_thread_history
  returns nothing when the application never set one, which
  is not the same as a conversation with no turns.
Never state that data does not exist. State that none was
returned under the constraints above.
Write your answer in ASD-STE100 Simplified Technical
English. Use short sentences with one idea in each.

Do you need the MCP server at all?

Yes, and it is worth considering given the archive. LangSmith's REST API covers runs, prompts and datasets with the same API key, and is documented independently of this package.

The connector's advantage is tool descriptions detailed enough that an agent picks the right call unprompted, which is the same verbosity that makes it cost 5,458 tokens.

For a frozen package, the API is the surface that will keep changing with the product. The connector will not.

Connect the server for exploratory work on recent runs. Use the REST API where you need something the frozen package does not expose.

LangSmith MCP server setup

The package runs with uvx or pip and needs a LangSmith API key in LANGSMITH_API_KEY.

Add LANGSMITH_WORKSPACE_ID for a multi-workspace account, and LANGSMITH_ENDPOINT for self-hosted instances.

claude mcp add-json "langsmith" '{"command":"uvx",
  "args":["langsmith-mcp-server"],
  "env":{"LANGSMITH_API_KEY":"<your-key>"}}'

{"mcpServers":{"langsmith":{"command":"uvx",
  "args":["langsmith-mcp-server"],
  "env":{"LANGSMITH_API_KEY":"<your-key>"}}}}

[mcp_servers.langsmith]
command = "uvx"
args = ["langsmith-mcp-server"]
env = { LANGSMITH_API_KEY = "<your-key>" }

Every other client

Each block below is the configuration for one client, with the file path and the key that client expects.

{"mcpServers":{"langsmith":{"command":"uvx",
  "args":["langsmith-mcp-server"],
  "env":{"LANGSMITH_API_KEY":"<your-key>"}}}}

{"mcpServers":{"langsmith":{"command":"uvx",
  "args":["langsmith-mcp-server"],
  "env":{"LANGSMITH_API_KEY":"<your-key>"}}}}

{"servers":{"langsmith":{"command":"uvx",
  "args":["langsmith-mcp-server"],
  "env":{"LANGSMITH_API_KEY":"<your-key>"}}}}

{"mcpServers":{"langsmith":{"command":"uvx",
  "args":["langsmith-mcp-server"],
  "env":{"LANGSMITH_API_KEY":"<your-key>"}}}}

LANGSMITH_API_KEY=<your-key> \
LANGSMITH_ENDPOINT=https://langsmith.example.com \
uvx langsmith-mcp-server

Read-only access, permissions and security

• No tool writes

  Four of the fifteen are named like writes, but push_prompt, create_dataset, update_examples and run_experiment are labeled Documentation-only in the README and implemented as zero-argument functions that return a markdown page of SDK examples. They take no parameters and reach nothing.

• The naming is the hazard, not the behavior

  An operator reading the tool list reasonably assumes push_prompt pushes a prompt. It returns instructions for pushing one. Budget for the confusion, not for the write.

• Nothing here spends money

  run_experiment returns documentation, so an agent calling it repeatedly costs context and nothing else.

• The archive does not remove the key's reach

  An archived repository is still an installed package holding an API key with your workspace's permissions.

Troubleshooting

No runs found in a project you know has traffic

Check the project name and the workspace. LANGSMITH_WORKSPACE_ID scopes results, and a query against the wrong workspace returns an empty result with no explanation.

A conversation has no thread history

get_thread_history needs a thread identifier that your application sets. Without one there is nothing to group, and the response looks the same as an empty conversation.

401 on connect

LANGSMITH_API_KEY is missing or belongs to another workspace. For a self-hosted instance, set LANGSMITH_ENDPOINT as well.

You want a fix that is not in 0.1.1

The repository is archived, so no further releases are expected. The LangSmith REST API is the surface that will keep changing with the product.

LangSmith MCP server: Reference

What engineers report

LangSmith is widely used for agent observability. The connector's repository carries 131 stars and 11 releases on PyPI, and it was archived on 14 August 2026. The signals below are repository state and package cadence rather than opinions.

“exceeds maximum allowed tokens (25000). Please use pagination, filtering, or limit parameters to reduce the response size.”

Repository facts from the GitHub API and package facts from PyPI, both retrieved 21 August 2026. Tool counts come from the server's own tools/list response, which agrees with the README's table.

Should you connect an agent to LangSmith?

Yes, though it is no longer actively maintained. Best for LangSmith teams who pin the version, allowlist read tools, and keep the REST API in reserve.

Strongest fit

Teams already on LangSmith who want an agent reading runs, prompts and experiments without leaving the editor.

Main advantage

Reach across the whole LangSmith model, including billing usage, which most agent-observability connectors omit.

Main weakness

The repository is archived and the package sits at 0.1.1, so what it exposes is what it will continue to expose.

Operational risk

There are no writes to gate. The four tools that read as writes return documentation, so the connector is read-only in practice even though nothing advertises it as such.

Frequently asked questions