This page was a little difficult to pass and there were no link
references to other relevant pages. Updated with improvements. No
technical details changed.

…ote (langchain-ai#4783)

Fixes DOC-1347

## Summary

- Updated section heading in the Python changelog from `` `deepagents`
v0.7.0 `` to `` `deepagents` v0.7.0a6 ``
- Removed the `<Note>` at the bottom of that section that stated the
changes are only available in the `deepagents` Python SDK

## Links

- Linear:
https://linear.app/langchain/issue/DOC-1347/update-deepagents-changelog-heading-to-v070a6-and-remove-sdk-only-note
- Slack:
https://langchain.slack.com/archives/C09G1T60QV9/p1783525164188249

## Verification

Not run; docs-only copy change.

## Reviewers

Requested review from: @npentrel, @lnhsingh

Co-authored-by: Docs Bot <brace@langchain.dev>

…i#4785)

…ain-ai#4784)

Fixes DOC-1348

## Summary

- Updated the Prerequisites section in `deployment-quickstart.mdx` to
mark Docker (and Docker Buildx on Apple Silicon) as optional rather than
required
- Added a note that `langgraph deploy` automatically falls back to a
remote build if Docker is not installed locally
- This aligns the quickstart with the CLI reference page, which already
correctly describes Docker as optional for `langgraph deploy`

## Links

- Linear:
https://linear.app/langchain/issue/DOC-1348/remove-docker-as-hard-prerequisite-from-langgraph-deploy-quickstart
- Slack:
https://langchain.slack.com/archives/C09G1T60QV9/p1783520934801329

## Verification

Not run; docs-only copy change.

## Reviewers

Requested review from: @katmayb, @fjmorris

---------

Co-authored-by: LangSmith Fleet <lauren@langchain.dev>
Co-authored-by: Kathryn May <44557882+katmayb@users.noreply.github.com>

- Added evaluator-spend.mdx.
- Modified docs.json & other pages, added links to evaluator-spend.mdx.

Fixes DOC-998

Fixes DOC-1354

Pins local docs development to Node 22 so `docs dev` uses a
Mintlify-supported LTS runtime instead of failing under Node 25+.

This also documents the distinction between the Python `docs` CLI and
the Mintlify `mint` CLI, and adds a small install hint for shells that
need to be relaunched before `.venv/bin/docs` is on `PATH`.

…rements (langchain-ai#4781)

## Summary
- Add a "Migrate with an AI agent" section to the SmithDB SDK migration
page with a copy/paste prompt clients can hand to a coding agent to
perform the migration
- Add a "Minimum SDK version" section consolidating the per-language
version requirements (Python `langsmith>=0.9.8`, TypeScript
`langsmith>=0.7.15`, Java `langsmith-java 0.1.0-beta.11`, Go
`langsmith-go v0.17.0`) that were previously scattered as inline
`<Note>` callouts across the runs-query and runs-retrieve snippets
- Remove the now-redundant inline version notes from `runs-query.mdx`
and `runs-retrieve.mdx`

)

…1 directory (langchain-ai#4809)

Bumps the uv group with 1 update in the / directory:
[soupsieve](https://github.com/facelessuser/soupsieve).

Updates `soupsieve` from 2.8 to 2.8.4
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/facelessuser/soupsieve/releases">soupsieve's
releases</a>.</em></p>
<blockquote>
<h2>2.8.4</h2>
<ul>
<li><strong>FIX</strong>: Fix another inefficient attribute pattern (<a
href="https://github.com/mauriceng98"><code>@​mauriceng98</code></a>).</li>
<li><strong>FIX</strong>: Limit total number of selectors processed in a
pattern to prevent massive selector requests (<a
href="https://github.com/mauriceng98"><code>@​mauriceng98</code></a>).</li>
</ul>
<h2>2.8.3</h2>
<ul>
<li><strong>FIX</strong>: Fix inefficient attribute pattern.</li>
</ul>
<h2>2.8.2</h2>
<ul>
<li><strong>FIX</strong>: Ensure custom selectors or namespace
dictionaries reject non-string keys (<a
href="https://github.com/mundanevision20"><code>@​mundanevision20</code></a>).</li>
<li><strong>FIX</strong>: Fix handling of <code>:in-range</code> and
<code>:out-of-range</code> with end of year weeks (<a
href="https://github.com/mundanevision20"><code>@​mundanevision20</code></a>).</li>
<li><strong>FIX</strong>: Fix a potential infinite loop in the pretty
printing debug function (<a
href="https://github.com/mundanevision20"><code>@​mundanevision20</code></a>).</li>
</ul>
<h2>2.8.1</h2>
<ul>
<li><strong>FIX</strong>: Changes in tests to accommodate latest Python
HTML parser changes.</li>
</ul>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/28108ab805818c832d9568142a99844fd95a0d39"><code>28108ab</code></a>
Limit excessive selectors</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/ef188721d6cc95641e99297b3a26ac17b7dfcfa7"><code>ef18872</code></a>
Fix test for Windows</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/eb4397618709186c109400448c6043b728217dc3"><code>eb43976</code></a>
Merge commit from fork</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/3a661b23b20e92b49b4683f0227c26c9d765267f"><code>3a661b2</code></a>
Fix typo in pseudo-classes.md (<a
href="https://redirect.github.com/facelessuser/soupsieve/issues/294">#294</a>)</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/0cb533d83bfc445c7c6321742a21f612305fc8ba"><code>0cb533d</code></a>
Update hatchling version requirement in pyproject.toml (<a
href="https://redirect.github.com/facelessuser/soupsieve/issues/290">#290</a>)</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/5aedc4180468e724aff46ddb3f738d35a3a9f724"><code>5aedc41</code></a>
Update doc theme</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/d7c47842a4f8d168243af96b62b9ad7fb84a2038"><code>d7c4784</code></a>
Attribute pattern fix (<a
href="https://redirect.github.com/facelessuser/soupsieve/issues/289">#289</a>)</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/09e106dc0fd6579c0327801fdcceb380ff60170e"><code>09e106d</code></a>
Fix grammar</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/09b27696ada6f07523a077950ce73da45579d524"><code>09b2769</code></a>
Update docs</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/c6e80fcab9ca4d3eaa61913778263e82bcecca1f"><code>c6e80fc</code></a>
Various fixes by <a
href="https://github.com/mundanevision20"><code>@​mundanevision20</code></a>
(<a
href="https://redirect.github.com/facelessuser/soupsieve/issues/288">#288</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/facelessuser/soupsieve/compare/2.8...2.8.4">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=soupsieve&package-manager=uv&previous-version=2.8&new-version=2.8.4)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore <dependency name> major version` will close this
group update PR and stop Dependabot creating any more for the specific
dependency's major version (unless you unignore this specific
dependency's major version or upgrade to it yourself)
- `@dependabot ignore <dependency name> minor version` will close this
group update PR and stop Dependabot creating any more for the specific
dependency's minor version (unless you unignore this specific
dependency's minor version or upgrade to it yourself)
- `@dependabot ignore <dependency name>` will close this group update PR
and stop Dependabot creating any more for the specific dependency
(unless you unignore this specific dependency or upgrade to it yourself)
- `@dependabot unignore <dependency name>` will remove all of the ignore
conditions of the specified dependency
- `@dependabot unignore <dependency name> <ignore condition>` will
remove the ignore condition of the specified dependency and ignore
conditions
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/docs/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

## Summary

Fixes an incorrect limit value in the LangSmith dashboards documentation
for the "Group by" feature.

**Change:** Updated the documented Group by limit from "top 5 elements"
to "top 20 elements" in `src/langsmith/dashboards.mdx`.

## Context

Reported by Marta Stangierska via
[#docs-plz](https://langchain.slack.com/archives/C09G1T60QV9/p1783603511705219):

> "Note that group by is limited to the top 5 elements by frequency."
>
> It seems like the limit is actually 20 now.

## Files changed

- `src/langsmith/dashboards.mdx` — Updated "top 5 elements" → "top 20
elements" in the Group by description (line 69)

## Linear issue


https://linear.app/langchain/issue/DOC-1357/fix-group-by-limit-in-langsmith-docs-5-20

## Slack message

https://langchain.slack.com/archives/C09G1T60QV9/p1783603511705219

## Verification

- [ ] Confirm the new limit of 20 is accurate with the LangSmith AI
Observability team
- [ ] Spot-check the rendered dashboards.mdx page once deployed

---------

Co-authored-by: Docs Bot <docs-bot@langchain.dev>
Co-authored-by: Kathryn May <44557882+katmayb@users.noreply.github.com>

…chain-ai#4187)

## Summary

Adds Terraform deployment guides for LangSmith self-hosted across AWS,
GCP, and Azure, mirroring the structure of the existing Helm setup
guide. The set is one shared overview (`self-host-terraform`) plus five
pages per provider:

- **Deploy**: end-to-end walkthrough from prerequisites to a running
install.
- **Architecture**: platform layers, managed services, networking,
ingress, secret flow, and the Terraform module graph.
- **Variables**: exhaustive reference for the `infra` (foundation) layer
inputs.
- **Quick reference**: condensed commands and add-on toggles.
- **Troubleshooting**: failure symptoms, causes, and fixes.

OpenShift is not included. The OCP module is still Preview upstream, so
it is left out until it matures.

## Why

Self-hosted customers using the public Terraform modules at
github.com/langchain-ai/terraform have been pointed at provider READMEs.
This PR brings those walkthroughs into the official docs so the
Terraform path matches the experience the Helm path already has.

## Accuracy audit

Every page was audited line by line against the actual modules in
github.com/langchain-ai/terraform (cloned as read-only reference).
Substantive corrections include:

- **Variables pages rewritten to be exhaustive** for each provider's
`infra` layer, sourced from the real `variables.tf` defaults and
validation blocks rather than a curated subset.
- **Provider secret models documented as they actually differ**, not
homogenized:
- AWS syncs secrets from SSM Parameter Store into the cluster through
the External Secrets Operator.
- GCP writes the Kubernetes Secrets directly from Terraform; Secret
Manager is an optional durable copy. No External Secrets Operator is
installed. An earlier draft incorrectly described an ESO sync path on
GCP (inherited from the upstream module docs, which carry the same
error); this is now corrected.
  - Azure writes `secrets.auto.tfvars` and stores secrets in Key Vault.
- **Azure managed services corrected**: Azure Managed Redis
(`Microsoft.Cache/redisEnterprise`, TLS port 10000) replaces the retired
classic Azure Cache for Redis (port 6380); `amr_sku` replaces
`redis_capacity`; PostgreSQL default version is 14; the `ltree`
extension was removed from the enabled list (the module enables
`pgcrypto`, `btree_gin`, `pg_trgm`, `btree_gist`, and `citext`).
- **AWS sizing corrected**: removed a fabricated table claiming
`sizing_profile` drives instance types. Infrastructure capacity comes
from the `infra` variables directly; `sizing_profile` only selects the
Helm sizing overlay. EKS version updated to 1.33.
- **GCP secret names corrected** to the `-credentials` suffixed names
the chart expects, and blob storage reframed around the shipped native
GCS mode (Workload Identity, no HMAC keys) with the S3-compatible mode
noted as optional.
- **Terminology aligned**: "stages" and "layers" throughout, matching
the consolidated guides.

## Areas needing careful review

- The three variables pages have the largest surface area. They are
scoped to the `infra` foundation layer, which is what these guides
deploy.
- The provider secret-flow sections, since each provider genuinely
differs and the differences are easy to over-normalize.
- The GCP architecture secret-flow diagram and cluster-infrastructure
table, which the ESO correction touched.

## Test plan

- [x] `make lint_prose` passes (0 errors across the 11 audited terraform
pages).
- [ ] `make broken-links` passes in CI. It cannot run locally here
because Mintlify does not support Node 25+. The one new internal link
(`/langsmith/self-host-scale`) was verified to exist and to have a
`docs.json` nav entry.
- [ ] Mintlify preview renders all pages and navigation entries resolve.

## AI involvement

Drafted and audited in collaboration with Claude Code (Opus). The author
directed content, scope, and accuracy decisions; Claude verified claims
against the Terraform modules, rewrote the variables references, applied
the corrections above, and ran Vale lint cleanup.

---------

Co-authored-by: Kathryn May <kathryn@langchain.dev>

- Added engine-security.mdx.
- Modified docs.json, added engine-security.mdx.
- Modified engine-overview.mdx, added links to both engine-security.mdx
and engine-self-hosted.mdx.
- Modified regions-faq.mdx, added link to engine-security.mdx.

Fixes DOC-1296

… guide (langchain-ai#4805)

## Summary
- Add a dedicated **Exceptions** section (Python/TypeScript) to the
SmithDB SDK migration guide (`src/langsmith/smithdb-sdk-migration.mdx`),
mapping the legacy `langsmith.utils` exception classes to the new
Stainless-generated ones.
- Add a **"Handle a run that is not found"** example to
`runs-retrieve.mdx`
- Add note on the fact that python methods are now async
- Fix some code samples

## Notable implementation details
- Java/Kotlin and Go are unaffected by the exceptions rename — both SDKs
were already Stainless-generated before SmithDB, so `NotFoundException`
/ `*langsmith.Error` (checked via `errors.As`) are unchanged between the
old and new methods.

…langchain-ai#4821)

## Summary
- Update the TypeScript exceptions table on the SmithDB migration guide
- Add a cURL tab to the exceptions section

## Test plan
- [x] `make lint_prose` passes on the changed file