---
title: "LangSmith MCP server for AI agents - actionlake"
description: "Give Claude 530 LangSmith actions through one MCP server: orgs, sandboxes, workspaces. Your agent never holds the credential. 100,000 calls a month free."
canonical: https://actionlake.com/connectors/langsmith
updated: 2026-10-23
---

Connectors · AI

# LangSmith MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to LangSmith and the rest of your systems. It gives your agent 530 LangSmith actions (244 reads, 184 writes and 102 that wait for a confirmation) and holds the LangSmith credential itself, so the agent never does. It paces LangSmith's limits and records every call.

[Connect LangSmith free](https://actionlake.com/console?start=sandbox) [See pricing](https://actionlake.com/pricing)

100,000 free calls a month. No card, no email, no password: press Start a free sandbox on the next screen and keep the key it shows you.

`claude mcp add --transport http actionlake https://connect.actionlake.com/mcp`

the first call

```
do {
  action: "langsmith.datasets.list",
  intent: "check which LangSmith account is connected"
}
# Read Datasets · read · 1 upstream call
```

- 530 actions in one signed file
- 244 read look, never change
- 184 write create and update
- 102 wait for a yes destructive, confirmed first

Set up

## Connect LangSmith in three steps.

1. 01
   ### Add actionlake to Claude.
   Run the command once in Claude Code; its consent page can create a free sandbox: 100,000 calls a month, no card. In the Claude app, add a custom connector with the URL below and paste your key on the consent page. Creating a new sandbox from the Claude app is not yet tested by us.
   `https://connect.actionlake.com/mcp`
2. 02
   ### Connect LangSmith.
   API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead. actionlake stores what you paste write-only, sealed under your tenant's key.
   - `api_key` LangSmith API key: in LangSmith open Settings > API Keys and create a Personal Access Token (lsv2_pt_...) or a…
   - `host` Where your LangSmith lives: your host
   - `organization_id` Your LangSmith organization id
   - `workspace_id` Your LangSmith workspace id
   ```
   account { op: "connections.requirements", service: "langsmith" }
   # free: exactly what to paste, and what is found for you
   ```
3. 03
   ### Make the first call. Check before a write.
   Ask in plain words; Claude picks the action. Before anything changes, `check` validates the call and prices it, for free.
   ```
   do    { action: "langsmith.datasets.list" }
   # read datasets · read · 1 upstream call
   check { action: "langsmith.datasets.create", input: { … } }
   # free: inputs, permissions, cost and confirmations, nothing sent
   ```

LangSmith at a glance

**Sign-in**

API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead.

**You provide**

`api_key`

**Setup fields**

`host` `organization_id` `workspace_id`

**Limits we pace to**

600 calls a minute, per connection

**API covered**

LangSmith API (spec 0.1.0, OpenAPI 3.1.0)`api.smith.langchain.com`

**Built from**

LangSmith's documentation, read 23 October 2026[api.smith.langchain.com](https://api.smith.langchain.com/openapi.json) [docs.smith.langchain.com](https://docs.smith.langchain.com/reference)

**Service file**

Version 1.0.0, signed (Ed25519)

**Cached reads**

189 of 244 reads, for up to 5 minutes; a write clears what it affects

**Shared entities**

4, so ask can line LangSmith up with your other systems

**Data**

Processed and stored in Sydney, Australia. No US region yet

Capabilities

## What an AI agent can do in LangSmith.

530 actions in 63 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

- **244** read look, never change
- **184** write create and update
- **102** destructive wait for a confirmation

1. ### Orgs
   52
   19 read · 20 write · 13 destructive
   Get Current Active Org Members · On Payment Method Created · Create Customers And Get Stripe Setup Intent · Create Stripe Checkout Sessions Endpoint
2. ### Sandboxes
   39
   14 read · 18 write · 7 destructive
   Execute a sandbox command · Glob a sandbox filesystem · Grep a sandbox filesystem · Create a sandbox
3. ### Workspaces
   39
   18 read · 11 write · 10 destructive
   List Workspaces · Get Current Active Workspace Members · Create Workspace · Patch Workspace
4. ### Annotation queues
   30
   12 read · 12 write · 6 destructive
   Create Annotation Queue · Export Annotation Queue Archived Runs · Get Annotation Queues · Populate Annotation Queue
5. ### Tracer sessions
   26
   11 read · 10 write · 5 destructive
   Create Tracer Session · Get Tracing Project Prebuilt Dashboard · Read Tracer Sessions · [Beta] Create Insights Job
6. ### Datasets
   24
   10 read · 10 write · 4 destructive
   Read Datasets · Create Dataset · Read Dataset · Update Dataset
7. ### Charts
   23
   10 read · 9 write · 4 destructive
   Create Chart · Read Chart Preview · Read Charts · Read Single Chart
8. ### Run
   23
   14 read · 5 write · 4 destructive
   Generate Query For Runs · Group Runs · Stats Group Runs · Query Runs

**55 more areas · 274 actions**

- Public **17**
- Feedback **14**
- Examples **13**
- Bulk exports **12**
- Runs **12**
- Threads **12**
- MCP vendors **9**
- Optimization jobs **9**
- Repos **9**
- Access policies **8**
- Oauth **8**
- Tools **8**
- Evaluators **7**
- Issues agent **7**
- API key **6**
- Comments **6**
- Gateway policies **6**
- Prompt webhooks **6**
- Agents **5**
- Alert rules **5**
- Data planes **5**
- Features **5**
- Playground settings **5**
- Scim tokens **5**
- Tags **5**
- Feedback configs **4**
- Fleet secrets **4**
- Hub environments **4**
- Issues **4**
- Model price map **4**
- TTL settings **4**
- Usage limits **4**
- Commits **3**
- Directories **3**
- Fleet users **3**
- MCP **3**
- Ownerships **3**
- Service accounts **3**
- Audit logs **2**
- Info **2**
- Integrations **2**
- Productfeedback **2**
- Sessions **2**
- Settings **2**
- Usage **2**
- Ace **1**
- Backfills **1**
- Experiments **1**
- Fleet orgs **1**
- Fleet tenants **1**
- Fleet threads **1**
- Likes **1**
- Me **1**
- Organizations **1**
- Prompts **1**

Top actions

## The LangSmith API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 14 of LangSmith's 14 action names are shared with at least one other connector, so an agent learns them once.

| Action | What it does | Effect | Same name in |
| --- | --- | --- | --- |
| `langsmith.datasets.list` Read Datasets | Read Datasets. | read | `analytics.dataset.list` 28 connectors |
| `langsmith.workspaces.list` List Workspaces | List Workspaces. | read | `work.workspace.list` 50 connectors |
| `langsmith.workspaces_current_members_active.list` Get Current Active Workspace Members | Get Current Active Workspace Members. | read | `work.workspace_member.list` 26 connectors |
| `langsmith.datasets.create` Create Dataset | Create Dataset. | write | `analytics.dataset.create` 21 connectors |
| `langsmith.workspaces.create` Create Workspace | Create Workspace. | write | `work.workspace.create` 36 connectors |
| `langsmith.datasets.delete` Delete Datasets | Delete Datasets. | destructive | `analytics.dataset.delete` 23 connectors |
| `langsmith.audit_logs.list` Get Audit Logs | Get Audit Logs. | read | `identity.audit_event.list` 51 connectors |
| `langsmith.service_accounts.list` Get Service Accounts | Get Service Accounts. | read | `identity.service_account.list` 35 connectors |
| `langsmith.service_accounts.create` Create Service Account | Create Service Account. | write | `identity.service_account.create` 32 connectors |
| `langsmith.orgs_current_members_active.list` Get Current Active Org Members | Get Current Active Org Members. | read | this connector |
| `langsmith.repos.fork` Fork Repo | Fork Repo. | write | this connector |
| `langsmith.repos.list` List Repos | List Repos. | read | this connector |

Try it

## Prompts to paste into Claude.

Each one names the LangSmith actions it will call, so you know what it reads and what it changes before you send it.

1. Read read
   Who has access in LangSmith? List every user and their role.
   calls `langsmith.orgs_current_members_active.list`
2. Write write
   Create Dataset in LangSmith. Show me exactly what you will send before you save it.
   calls `langsmith.datasets.create`

Governance for LangSmith

## Destructive actions wait for a yes.

102 LangSmith actions are destructive. The first 10:

- Delete Datasets `langsmith.datasets.delete`
- Delete Workspace `langsmith.workspaces.delete`
- Delete Service Account `langsmith.service_accounts.delete`
- Add Members To Current Org Batch `langsmith.orgs_current_members.batch`
- Add Member To Current Org `langsmith.orgs_current_members.create`
- Add Basic Auth Members To Current Org `langsmith.orgs_current_members_basic.batch`
- Add Members To Current Workspace Batch `langsmith.workspaces_current_members.batch`
- Add Member To Current Workspace `langsmith.workspaces_current_members.create`
- Attach access policies to a role `langsmith.platform_orgs_current_roles_access_policies.attach`
- Attach access policies to a role `langsmith.platform_orgs_current_access_policies.attach_to_role`

A destructive action runs only with `confirm: true`. Under the `human` policy the agent cannot confirm for itself: the run stops until a named person approves, in the console or from a signed link in an email. The cap on a key is refused before the call is sent, never after.

do langsmith.datasets.delete confirm_policy: **human**

Delete Datasets?

**intent:** the user asked me to delete Datasets

run suspended · waiting for a named person · key _langsmith-agent_ cannot confirm itself

In the console, or from an emailed signed link.

a key for this connector

```
account {
  op: "keys.create", name: "langsmith-agent",
  allowed_actions: ["langsmith.*"], cap_calls: 2000,
  confirm_policy: "human_for_writes",
  approval_rules: [
    { match: "langsmith.datasets.delete", approver: "lead@example.com" },
    { match: "langsmith.workspaces.delete", approver: "lead@example.com" },
    { match: "langsmith.service_accounts.delete", approver: "lead@example.com" }
  ]
}
```

Illustrative. Scoped to LangSmith only, capped at 2,000 upstream calls, every write confirmed, and the destructive ones approved by a named person.

Cost

## What LangSmith costs on actionlake.

Every LangSmith action is one upstream call, the only thing actionlake bills. Searching, `check`, `cost`, `account` and any call LangSmith refused with 401, 403, 429 or a 5xx are free. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

- 1 upstream call per action all 530 LangSmith actions
- 530 actions page each page of a long list is one call
- 189 reads can be cached a cache hit costs nothing
- 100,000 calls a month, free paid plans from US$49 a month for 250,000

[See pricing](https://actionlake.com/pricing)

Action index

## Every LangSmith action.

530 actions by area: the id your agent calls, what it does, and its effect.

**Orgs 52**

- `langsmith.orgs_current.payment_method` On Payment Method Created write
- `langsmith.orgs_current.setup` Create Customers And Get Stripe Setup Intent write
- `langsmith.orgs_current.stripe_checkout_session` Create Stripe Checkout Sessions Endpoint write
- `langsmith.orgs_current_members_active.list` Get Current Active Org Members read
- `langsmith.orgs_current.get` Get Organization Info read
- `langsmith.orgs_current.plan` Change Payment Plan destructive
- and 46 more in this area

**Sandboxes 39**

- `langsmith.sandboxes.execute` Execute a sandbox command write
- `langsmith.sandboxes.glob` Glob a sandbox filesystem read
- `langsmith.sandboxes.grep` Grep a sandbox filesystem read
- `langsmith.sandboxes_boxes.create` Create a sandbox write
- `langsmith.sandboxes_boxes.batch_delete` Batch delete sandboxes destructive
- `langsmith.sandboxes_boxes.delete` Delete a sandbox destructive
- and 33 more in this area

**Workspaces 39**

- `langsmith.workspaces.create` Create Workspace write
- `langsmith.workspaces.list` List Workspaces read
- `langsmith.workspaces.update` Patch Workspace write
- `langsmith.workspaces_current_members_active.list` Get Current Active Workspace Members read
- `langsmith.workspaces.delete` Delete Workspace destructive
- `langsmith.workspaces.get` Get Workspace read
- and 33 more in this area

**Annotation queues 30**

- `langsmith.annotation_queues.create` Create Annotation Queue write
- `langsmith.annotation_queues.export` Export Annotation Queue Archived Runs write
- `langsmith.annotation_queues.list` Get Annotation Queues read
- `langsmith.annotation_queues.populate` Populate Annotation Queue write
- `langsmith.annotation_queues.delete` Delete Annotation Queues destructive
- `langsmith.annotation_queues.delete_annotation_queue` Delete Annotation Queue destructive
- and 24 more in this area

**Tracer sessions 26**

- `langsmith.sessions.create` Create Tracer Session write
- `langsmith.sessions.dashboard` Get Tracing Project Prebuilt Dashboard read
- `langsmith.sessions.list` Read Tracer Sessions read
- `langsmith.sessions_insights.create` [Beta] Create Insights Job write
- `langsmith.sessions.delete` Delete Tracer Session destructive
- `langsmith.sessions.delete_tracer_sessions` Delete Tracer Sessions destructive
- and 20 more in this area

**Datasets 24**

- `langsmith.datasets.create` Create Dataset write
- `langsmith.datasets.get` Read Dataset read
- `langsmith.datasets.list` Read Datasets read
- `langsmith.datasets.update` Update Dataset write
- `langsmith.datasets.clone` Clone Dataset write
- `langsmith.datasets.comparative` Create Comparative Experiment write
- and 18 more in this area

**Charts 23**

- `langsmith.charts.create` Create Chart write
- `langsmith.charts.preview` Read Chart Preview read
- `langsmith.charts.query` Read Charts read
- `langsmith.charts.query_one` Read Single Chart read
- `langsmith.charts.delete` Delete Chart destructive
- `langsmith.charts.update` Update Chart write
- and 17 more in this area

**Run 23**

- `langsmith.runs.generate_query` Generate Query For Runs read
- `langsmith.runs.group` Group Runs read
- `langsmith.runs.group_stats` Stats Group Runs read
- `langsmith.runs.query` Query Runs read
- `langsmith.runs.delete` Delete Runs destructive
- `langsmith.runs.delete_traces` Delete Runs Abac destructive
- and 17 more in this area

**Public 17**

- `langsmith.public.query_runs` Query Shared Runs read
- `langsmith.public_datasets_comparative.list` Read Shared Comparative Experiments read
- `langsmith.public_datasets_feedback.list` Read Shared Dataset Feedback read
- `langsmith.public_datasets_runs.generate_query` Generate Query For Shared Dataset Runs read
- `langsmith.public.get_run` Get Shared Run By Id read
- `langsmith.public.get_shared_run` Get Shared Run read
- and 11 more in this area

**Feedback 14**

- `langsmith.feedback.create` Create Feedback write
- `langsmith.feedback.list` Read Feedbacks read
- `langsmith.feedback_formulas.create` Create Feedback Formula Ep write
- `langsmith.feedback_formulas.list` List Feedback Formula Ep read
- `langsmith.feedback.delete` Delete Feedback destructive
- `langsmith.feedback.get` Read Feedback read
- and 8 more in this area

**Examples 13**

- `langsmith.examples.bulk` Create Examples write
- `langsmith.examples.create` Create Example write
- `langsmith.examples.list` Read Examples read
- `langsmith.examples.validate` Validate Example read
- `langsmith.examples.delete` Delete Example destructive
- `langsmith.examples.delete_examples` Delete Examples destructive
- and 7 more in this area

**Bulk exports 12**

- `langsmith.bulk_exports.create` Create Bulk Export write
- `langsmith.bulk_exports.list` Get Bulk Exports read
- `langsmith.bulk_exports_destinations.create` Create Bulk Export Destination write
- `langsmith.bulk_exports_destinations.list` Get Bulk Export Destinations read
- `langsmith.bulk_exports.get` Get Bulk Export read
- `langsmith.bulk_exports.update` Cancel Bulk Export write
- and 6 more in this area

**Runs 12**

- `langsmith.public.query_runs_v2` Query public shared trace runs read
- `langsmith.runs.create` Create a Run write
- `langsmith.runs.ingest_batch` Ingest Runs (Batch JSON) write
- `langsmith.runs.query_v2` Query runs read
- `langsmith.public.get_run_v2` Get a public shared trace run read
- `langsmith.runs.get_a_single_run` Get a single run read
- and 6 more in this area

**Threads 12**

- `langsmith.public_threads_traces.list` List traces in a public shared thread read
- `langsmith.threads.query` Query Threads read
- `langsmith.threads.stats` Query Thread Stats read
- `langsmith.threads_share.create` Share a thread write
- `langsmith.public_threads.get` Get a public shared thread's manifest read
- `langsmith.public_threads_runs.get` Get one run in a public shared thread read
- and 6 more in this area

**MCP vendors 9**

- `langsmith.platform_mcp_vendors.get` List MCP vendors read
- `langsmith.platform_mcp_vendors_mcp_servers.list` List MCP servers for a vendor read
- `langsmith.platform_mcp_vendors_settings.create` Create vendor settings write
- `langsmith.platform_mcp_vendors_tools.list` List tools for a vendor read
- `langsmith.platform_mcp_vendors.get_mcp_vendor` Get MCP vendor read
- `langsmith.platform_mcp_vendors_account.get` Get vendor account read
- and 3 more in this area

**Optimization jobs 9**

- `langsmith.repos_optimization_jobs.create` Create Job write
- `langsmith.repos_optimization_jobs.list` List Jobs read
- `langsmith.repos_optimization_jobs_logs.create` Create Log write
- `langsmith.repos_optimization_jobs_logs.list` List Job Logs read
- `langsmith.repos_optimization_jobs.delete` Delete Job destructive
- `langsmith.repos_optimization_jobs.get` Get Job read
- and 3 more in this area

**Repos 9**

- `langsmith.repos.create` Create Repo write
- `langsmith.repos.fork` Fork Repo write
- `langsmith.repos.list` List Repos read
- `langsmith.repos.optimize_job` Optimize Prompt Job write
- `langsmith.repos.delete` Delete Repos destructive
- `langsmith.repos.delete_repo` Delete Repo destructive
- and 3 more in this area

**Access policies 8**

- `langsmith.platform_orgs_current_access_policies.attach_to_role` Attach access policies to a role destructive
- `langsmith.platform_orgs_current_access_policies.create` Create an access policy write
- `langsmith.platform_orgs_current_access_policies.get` List access policies read
- `langsmith.platform_orgs_current_roles_access_policies.attach` Attach access policies to a role destructive
- `langsmith.platform_orgs_current_access_policies.delete` Delete an access policy destructive
- `langsmith.platform_orgs_current_access_policies.get_an_access_policy` Get an access policy read
- and 2 more in this area

**Oauth 8**

- `langsmith.platform_oauth_authorized_apps.list` List authorized applications read
- `langsmith.platform_oauth_clients.create` Create an OAuth client write
- `langsmith.platform_oauth_clients.get` List OAuth clients read
- `langsmith.platform_oauth_clients.get_an_oauth_client` Get an OAuth client read
- `langsmith.platform_oauth_authorized_apps.delete` Revoke an authorized application destructive
- `langsmith.platform_oauth_clients.delete` Delete an OAuth client destructive
- and 2 more in this area

**Tools 8**

- `langsmith.platform_tools.create` Create a tool write
- `langsmith.platform_tools.get` Get a tool by handle read
- `langsmith.platform_tools.list` List tools read
- `langsmith.platform_tools.update` Update a tool by handle write
- `langsmith.platform_tools.delete` Delete a tool by handle destructive
- `langsmith.platform_tools_id.delete` Delete a tool by ID destructive
- and 2 more in this area

**Evaluators 7**

- `langsmith.platform_evaluators.create` Create evaluator write
- `langsmith.platform_evaluators.get` Get evaluator read
- `langsmith.platform_evaluators.list` List evaluators read
- `langsmith.platform_evaluators.update` Update evaluator write
- `langsmith.platform_evaluators.delete` Bulk delete evaluators destructive
- `langsmith.platform_evaluators.delete_evaluator` Delete evaluator destructive
- and 1 more in this area

**Issues agent 7**

- `langsmith.platform_issues_agent.list` [Beta] List issues agent configs read
- `langsmith.platform_sessions_issues_agent.create` [Beta] Create the issues agent for a session write
- `langsmith.platform_sessions_issues_agent.get` [Beta] Get the issues agent config for a session read
- `langsmith.platform_sessions_issues_agent.update` [Beta] Update the issues agent config for a session write
- `langsmith.platform_sessions_issues_agent.delete` [Beta] Delete the issues agent for a session destructive
- `langsmith.platform_sessions_issues_agent_overview.update` [Beta] Save the agent overview for a session write
- and 1 more in this area

**API key 6**

- `langsmith.api_key.create` Generate Api Key write
- `langsmith.api_key.list` Get Api Keys read
- `langsmith.api_key_current.create` Generate Personal Access Token write
- `langsmith.api_key_current.list` Get Personal Access Tokens read
- `langsmith.api_key.delete` Delete Api Key destructive
- `langsmith.api_key_current.delete` Delete Personal Access Token destructive

**Comments 6**

- `langsmith.comments.create` Create Comment write
- `langsmith.comments.create_sub_comment` Create Sub Comment write
- `langsmith.comments.get_sub_comments` Get Sub Comments read
- `langsmith.comments.like` Like Comment write
- `langsmith.comments.list` Get Comments read
- `langsmith.comments_like.delete` Unlike Comment destructive

**Gateway policies 6**

- `langsmith.platform_gateway_policies.create` Create a gateway policy write
- `langsmith.platform_gateway_policies.get` Get a gateway policy read
- `langsmith.platform_gateway_policies.list` List gateway policies read
- `langsmith.platform_gateway_policies.search` Search gateway policies by subject value set read
- `langsmith.platform_gateway_policies.delete` Delete a gateway policy destructive
- `langsmith.platform_gateway_policies.update` Update a gateway policy write

**Prompt webhooks 6**

- `langsmith.prompt_webhooks.create` Create Prompt Webhook write
- `langsmith.prompt_webhooks.get` Get Prompt Webhook read
- `langsmith.prompt_webhooks.list` List Prompt Webhooks read
- `langsmith.prompt_webhooks.test` Test Prompt Webhook read
- `langsmith.prompt_webhooks.delete` Delete Prompt Webhook destructive
- `langsmith.prompt_webhooks.update` Update Prompt Webhook write

**Agents 5**

- `langsmith.agents.create` Create an Agent write
- `langsmith.agents.get` Get an Agent read
- `langsmith.agents.list` List Agents read
- `langsmith.agents.update` Update an Agent write
- `langsmith.agents.delete` Delete an Agent destructive

**Alert rules 5**

- `langsmith.platform_alerts.create` Create an alert rule write
- `langsmith.platform_alerts.get` Get an alert rule read
- `langsmith.platform_alerts.test` Test an alert action to determine if configuration is valid read
- `langsmith.platform_alerts.update` Update an alert rule write
- `langsmith.platform_alerts.delete` Delete an alert rule destructive

**Data planes 5**

- `langsmith.orgs_current_data_planes.create` Create a new data plane write
- `langsmith.orgs_current_data_planes.get` List data planes for the current organization read
- `langsmith.orgs_current_data_planes.get_a_data_plane` Get a data plane read
- `langsmith.orgs_current_data_planes.update` Update data plane settings write
- `langsmith.orgs_current_data_planes.delete` Delete a data plane destructive

**Features 5**

- `langsmith.platform_features.list` List feature configurations read
- `langsmith.platform_features_default_model.delete` Delete default model for a feature destructive
- `langsmith.platform_features_default_model.update` Set default model for a feature write
- `langsmith.platform_features_disabled_models.update` Disable a model for a feature write
- `langsmith.platform_features_disabled_models.delete` Re-enable a disabled model for a feature destructive

**Playground settings 5**

- `langsmith.playground_settings.create` Create Playground Settings write
- `langsmith.playground_settings.get` Get Playground Settings read
- `langsmith.playground_settings.list` List Playground Settings read
- `langsmith.playground_settings.update` Update Playground Settings write
- `langsmith.playground_settings.delete` Delete Playground Settings destructive

**Scim tokens 5**

- `langsmith.platform_orgs_current_scim_tokens.create` Create a SCIM token write
- `langsmith.platform_orgs_current_scim_tokens.get` Get a SCIM token read
- `langsmith.platform_orgs_current_scim_tokens.list` List SCIM tokens read
- `langsmith.platform_orgs_current_scim_tokens.update` Update a SCIM token write
- `langsmith.platform_orgs_current_scim_tokens.delete` Delete a SCIM token destructive

**Tags 5**

- `langsmith.repos_tags.create` Create Tag write
- `langsmith.repos_tags.get` Get Tag read
- `langsmith.repos_tags.get_tags` Get Tags read
- `langsmith.repos_tags.update` Update Tag write
- `langsmith.repos_tags.delete` Delete Tag destructive

**Feedback configs 4**

- `langsmith.feedback_configs.create` Create Feedback Config Endpoint write
- `langsmith.feedback_configs.delete` Delete Feedback Config Endpoint destructive
- `langsmith.feedback_configs.list` List Feedback Configs Endpoint read
- `langsmith.feedback_configs.update` Update Feedback Config Endpoint write

**Fleet secrets 4**

- `langsmith.fleet_secrets.create` Bulk set or delete workspace secrets write
- `langsmith.fleet_secrets.delete` Delete a workspace secret destructive
- `langsmith.fleet_secrets.list` List workspace secret names read
- `langsmith.fleet_secrets.update` Set a workspace secret write

**Hub environments 4**

- `langsmith.hub_environments.create` Create hub environments model write
- `langsmith.hub_environments.delete` Delete hub environments model destructive
- `langsmith.hub_environments.get` List hub environments read
- `langsmith.hub_environments.update` Update hub environments model write

**Issues 4**

- `langsmith.platform_issues.get` [Beta] Get issue read
- `langsmith.platform_issues.list` [Beta] List issues read
- `langsmith.platform_issues.views` [Beta] Mark issue viewed write
- `langsmith.platform_sessions_issues_views.get` [Beta] List viewed issues for a session read

**Model price map 4**

- `langsmith.model_price_map.create` Create New Model Price write
- `langsmith.model_price_map.delete` Delete Model Price destructive
- `langsmith.model_price_map.list` Read Model Price Map read
- `langsmith.model_price_map.update` Update Model Price write

**TTL settings 4**

- `langsmith.ttl_settings.list` List Ttl Settings read
- `langsmith.ttl_settings.update` Upsert Ttl Settings write
- `langsmith.workspaces_current_ttl_settings.get` Get workspace TTL settings read
- `langsmith.workspaces_current_ttl_settings.update` Update workspace TTL settings write

**Usage limits 4**

- `langsmith.usage_limits.delete` Delete Usage Limit destructive
- `langsmith.usage_limits.list` List Usage Limits read
- `langsmith.usage_limits.update` Upsert Usage Limit write
- `langsmith.usage_limits_org.list` List Org Usage Limits read

**Commits 3**

- `langsmith.commits.create` Create a commit write
- `langsmith.commits.get` Get a commit read
- `langsmith.commits.list` List commits read

**Directories 3**

- `langsmith.platform_hub_repos_directories.delete` Delete directory repository destructive
- `langsmith.platform_hub_repos_directories.get` Get directory contents read
- `langsmith.platform_hub_repos_directories_commits.create` Create directory commit write

**Fleet users 3**

- `langsmith.fleet_tenants_users.get` Get Fleet user in tenant read
- `langsmith.fleet_users.list` List Fleet users read
- `langsmith.fleet_users_current.get` Get current Fleet user read

**MCP 3**

- `langsmith.mcp_proxy.create` Proxy write
- `langsmith.mcp_tools.delete` Invalidate Tools Cache destructive
- `langsmith.mcp_tools.get` Get Tools read

**Ownerships 3**

- `langsmith.repos_owners.create` Add Repo Owner write
- `langsmith.repos_owners.delete` Remove Repo Owner destructive
- `langsmith.repos_owners.get` List Repo Owners read

**Service accounts 3**

- `langsmith.service_accounts.create` Create Service Account write
- `langsmith.service_accounts.delete` Delete Service Account destructive
- `langsmith.service_accounts.list` Get Service Accounts read

**Audit logs 2**

- `langsmith.audit_logs.list` Get Audit Logs read
- `langsmith.audit_logs_operations.get` List Audit Log Operations read

**Info 2**

- `langsmith.info.get` Get server info read
- `langsmith.info.health` Get Health Info read

**Integrations 2**

- `langsmith.agent_builder_integrations.get` Get Agent Builder integrations settings read
- `langsmith.agent_builder_integrations.update` Update Agent Builder integrations settings write

**Productfeedback 2**

- `langsmith.platform_product_feedbacks.create` [Alpha] Submit product feedback write
- `langsmith.platform_product_feedbacks.get` [Alpha] Get product feedback read

**Sessions 2**

- `langsmith.platform_sessions_agent_versions.list` List agent versions for a project read
- `langsmith.sessions_resolutions.get` [Beta] Resolve an address to its tracing project read

**Settings 2**

- `langsmith.settings.get` Get Settings read
- `langsmith.settings.handle` Set Tenant Handle write

**Usage 2**

- `langsmith.fleet_usage_agents.list` List Fleet agent costs read
- `langsmith.fleet_usage_users.list` List Fleet user costs read

**Ace 1**

- `langsmith.ace.execute` Execute write

**Backfills 1**

- `langsmith.platform_ops_backfills.restart` Restart a backfill job write

**Experiments 1**

- `langsmith.runs_experiments.evaluate` Evaluate Experiment Adhoc write

**Fleet orgs 1**

- `langsmith.fleet_orgs.list` List organizations read

**Fleet tenants 1**

- `langsmith.fleet_orgs_tenants.list` List tenants read

**Fleet threads 1**

- `langsmith.fleet_threads.sandbox_activation` Activate a thread sandbox write

**Likes 1**

- `langsmith.likes.create` Like Repo write

**Me 1**

- `langsmith.me_providers.get` Get the authenticated user's provider user ID read

**Organizations 1**

- `langsmith.platform_orgs_current_info.get` Get current organization info read

**Prompts 1**

- `langsmith.prompts.invoke_prompt` Invoke Prompt write

This index shows a sample of each area. All 530 actions, with their inputs and prices, are in the [catalogue](https://actionlake.com/console/catalogue), which a free sandbox key can search.

Questions

## The LangSmith MCP server, answered.

**Is there a LangSmith MCP server for Claude?**

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 530 LangSmith actions across orgs, sandboxes, workspaces and annotation queues, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by LangSmith.

**What do I need to connect LangSmith?**

LangSmith API key: in LangSmith open Settings > API Keys and create a Personal Access Token (lsv2_pt_...) or a…. Plus your host, your organization id and your workspace id. The free call account { op: "connections.requirements", service: "langsmith" } lists exactly what to paste.

**Can an AI agent delete or change things in LangSmith?**

Only if its key allows it, and only with an explicit confirmation. 102 LangSmith actions are destructive, including delete Datasets, delete Workspace and delete Service Account. Each runs only with confirm: true. Under the human confirm policy, or an approval rule on the key, the run waits until a named person approves it.

**Will an AI agent use up my LangSmith API limit?**

actionlake paces every call to the limits in the LangSmith service file: 600 calls a minute per connection. Repeated reads are answered from cache (189 LangSmith actions can be), which never reaches LangSmith.

**What does a LangSmith action cost?**

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (530 LangSmith actions page). Cache hits, searches, dry runs and calls LangSmith refused with 401, 403, 429 or a 5xx are free. The Free plan includes 100,000 calls a month with no card. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

**Which LangSmith API does it cover?**

The LangSmith API (spec 0.1.0, OpenAPI 3.1.0) at api.smith.langchain.com by default, or the base URL set when you connect. It is built from LangSmith's documentation as read on 23 October 2026. 530 actions in 63 areas, each one in the signed service file (version 1.0.0).

**Does actionlake keep my LangSmith data, and where?**

actionlake processes and stores data in Sydney, Australia; there is no US region yet. Reads may be cached for up to 5 minutes, and a write clears the reads it affects. Each run's request and response is kept, sealed, for your plan's trace window: 1 day on Free, 3 days on Starter, 7 days on Pro, 14 days on Team, 30 days on Business and 30 days on Enterprise. Your LangSmith credential is sealed under your tenant's key, never logged and never returned. Your traffic is not used to train models.

**Which AI clients can use it?**

Claude Code, tested by us end to end, and the Claude app as a custom connector, tested by us with an existing key. Other MCP clients that support remote servers over Streamable HTTP (MCP revisions 2025-06-18 and 2025-11-25) and OAuth dynamic client registration, or a bearer header, should connect the same way; we have not tested them. Or plain REST, under the same action names.

## More connectors like LangSmith

Shares the most action names with

- [OpenAI Admin 8](https://actionlake.com/connectors/openai-admin)
- [Mistral AI 7](https://actionlake.com/connectors/mistral-ai)
- [Portkey 5](https://actionlake.com/connectors/portkey)
- [Botpress 5](https://actionlake.com/connectors/botpress)
- [Fireworks AI 5](https://actionlake.com/connectors/fireworks-ai)
- [Langfuse 3](https://actionlake.com/connectors/langfuse)

Joins with, on shared entities

- [Preset 4](https://actionlake.com/connectors/preset)
- [Microsoft Fabric 3](https://actionlake.com/connectors/microsoft-fabric)
- [Baserow 3](https://actionlake.com/connectors/baserow)
- [PandaDoc 3](https://actionlake.com/connectors/pandadoc)
- [Adobe Workfront 2](https://actionlake.com/connectors/adobe-workfront)
- [Tableau 2](https://actionlake.com/connectors/tableau)

Browse

- [All AI connectors](https://actionlake.com/connectors/category/ai)
- [All connectors](https://actionlake.com/connectors)
- [How actionlake compares](https://actionlake.com/compare)

actionlake is independent. It is not made or endorsed by LangSmith; the name is the vendor's own. Counts are from the LangSmith service file, version 1.0.0.

## Give your agent LangSmith today.

[Connect LangSmith free](https://actionlake.com/console?start=sandbox) [See pricing](https://actionlake.com/pricing)

Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.

---

Source: https://actionlake.com/connectors/langsmith
More: https://actionlake.com/llms.txt