Conversation
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
WalkthroughAdds a new installation guide Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Suggested reviewers
Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 5
🤖 Fix all issues with AI agents
In `@docs/en/installation/kubeflow.mdx`:
- Line 176: Replace the hard-coded namespace string
"cluster.local/ns/kubeflow-new/sa/ml-pipeline-ui" with a clear placeholder (for
example "cluster.local/ns/<kubeflow-namespace>/sa/ml-pipeline-ui") or add a
brief clarifying sentence near that occurrence explaining what "kubeflow-new"
refers to (i.e., the Kubeflow installation namespace) so readers know to
substitute their environment-specific namespace; update the text around the
symbol to use the placeholder or the explanatory note consistently.
- Line 19: The line that references an internal Confluence link for obtaining
the platform Dex CA certificate must be replaced with publicly accessible
guidance: remove or replace the confluence.alauda.cn URL and either (1) inline
the necessary steps to obtain the platform Dex CA certificate (a short
step-by-step list explaining where to find/export the CA cert from the platform
and any CLI commands or UI paths), or (2) link to a public documentation page or
official upstream source that provides the same instructions; update the
sentence mentioning "platform dex CA certificate" accordingly so external
readers can follow the steps without access to internal resources.
- Around line 183-185: The docs use two inconsistent names for the same plugin —
"kftrainer" and "kftraining" — which will confuse users; pick one canonical
identifier and update the occurrences accordingly (or explicitly state that
"kftrainer" is the display name and "kftraining" is the package name). Edit the
text referencing kftrainer/kftraining in the Kubeflow section (lines referencing
"kftrainer", "kftraining", "KubeflowTrainingOperator", and "Cluster Plugins") so
all mentions either use the unified name or include a parenthetical
clarification like "kftrainer (display name) / kftraining (package name)" to
make the relationship clear to users searching the Cluster Plugins UI.
- Line 21: Single-line summary: Replace the non-actionable "please consult
relevant colleagues" with actionable guidance for ServiceMesh v2 configuration.
Update the sentence in the docs (the paragraph referencing Platform Management
-> Resource Management and the spec field for ServiceMesh v1) to either (a) link
to the ServiceMesh v2 configuration guide, (b) inline the v2 steps if short and
stable, or (c) specify the exact internal team or channel to contact (e.g.,
Platform Team docs or Slack channel), so external readers have a clear next
step; ensure the text references "ServiceMesh v1" and "ServiceMesh v2"
explicitly and that the link or contact replaces the current phrase.
- Around line 33-51: The PEM block beginning with "-----BEGIN CERTIFICATE-----"
and ending with "-----END CERTIFICATE-----" contains a real CA certificate;
replace that entire PEM block with a clearly labeled placeholder (e.g.,
"-----BEGIN CERTIFICATE-----\n<PLACEHOLDER CERTIFICATE FOR DOCUMENTATION
ONLY>\n-----END CERTIFICATE-----") and add a short inline note that this is a
dummy/example certificate for documentation purposes only so no real PKI
material is published.
🧹 Nitpick comments (2)
docs/en/installation/kubeflow.mdx (2)
122-131: Addyamllanguage identifier to fenced code blocks for syntax highlighting.This code block (and the one at line 137) contains YAML but lacks a language identifier, which means no syntax highlighting will be applied. Add
yamlto improve readability:-``` +```yaml apiVersion: kubeflow.org/v1beta1 kind: ProfileSame applies to the code block starting at line 137.
108-108: Consider using a placeholder instead of a hardcoded example IP.The redirect URI uses a hardcoded IP (
192.168.139.133:30665) that differs from the example platform address on line 82 (192.168.171.123). While these may intentionally refer to different clusters, using a placeholder like<your-cluster-address>:<nodeport>(with an explanatory note) would be clearer and more consistent.
docs/en/installation/kubeflow.mdx
Outdated
| -----BEGIN CERTIFICATE----- | ||
| MIIDKzCCAhOgAwIBAgIRAK9C9PuDXtYFvybudWQkN4UwDQYJKoZIhvcNAQELBQAw | ||
| EDEOMAwGA1UEChMFY3BhYXMwHhcNMjUwMzEwMDkxODAzWhcNMzUwMzA4MDkxODAz | ||
| WjASMRAwDgYDVQQKEwdrdWJlLWNhMIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIB | ||
| CgKCAQEAmChGjtwWOPvj0Ca3TkuPxxx6jg4oDTAPqyowT2pcaVeNhFwoMmCCkFXm | ||
| 7brFKXCc7IE1kHq5dbRCn+UwCA46g7zvz8b7SY/0qRymwTlYqRILDZacwWHUSJSD | ||
| cDyK297V+Ig5oIno6fTa2FWSJBqyxqivZ3lzf1XpsiwSPPXol+LclUne0fDiM98C | ||
| dBQWKDYadwlcluuPUHULthA3OjcKGpmyV7cyTHPcRjBSmkAmuL0bQhbWhkB8G9oe | ||
| 4cp2joo/qVsSzeUepkHeTD9PPk1AZ59FE8DDgL0FRREE7vou6g7fbOZL98pC4ldg | ||
| ZIY/EB5v38uR6J25uzLPFSf75vbwHwIDAQABo34wfDAOBgNVHQ8BAf8EBAMCBaAw | ||
| DAYDVR0TAQH/BAIwADAfBgNVHSMEGDAWgBQk8E8JWyAANbALLaeAxZ17adgq/TA7 | ||
| BgNVHREENDAyggVjcGFhc4ILZXhhbXBsZS5jb22HBH8AAAGHEAAAAAAAAAAAAAAA | ||
| AAAAAAGHBMCoq/MwDQYJKoZIhvcNAQELBQADggEBAIXo0V2jMeRd4cw5p3FWoFno | ||
| VWno7Cy7ENvVjgfQymcWbGi6fXWvkDBUPCmqv5bosUVyAOJ/p92g861nCAo3jxoZ | ||
| voCTDN4xU+t0xs2hMTKHsSB7v3n18rBtqcVpUvm1it/NyeOU4HiYfPTPkRVugGf4 | ||
| gtYknrU6Skt9BkiNy+2Jcsb6V3mAJ5GQzbT0qPL1vKWkBB9oCbjMwJggsW+TdKgY | ||
| KJuII0m6JNDUlKLCazLL8OvXq84Nu+cJ6QaNOT0gBRIWSPA+UbAsibbFnf0VOeeU | ||
| WforZLredR6GKc2qMdKdcW4G+8fRSWcx0gEIRquoQH1P7yIEJ3xOGoxQfIRVpls= | ||
| -----END CERTIFICATE----- |
There was a problem hiding this comment.
Embedded certificate appears to be a real CA certificate — use a placeholder instead.
This PEM block looks like an actual internal CA certificate rather than a placeholder. Publishing real certificates in documentation (especially in a public repo) is a security concern — it leaks internal PKI details and could aid targeted attacks. Replace it with a clearly marked placeholder, e.g.:
- path: spec.values.pilot.jwksResolverExtraRootCA
value: |
-----BEGIN CERTIFICATE-----
- MIIDKzCCAhOgAwIBAgIRAK9C9PuDXtYFvybudWQkN4UwDQYJKoZIhvcNAQELBQAw
- ...
- WforZLredR6GKc2qMdKdcW4G+8fRSWcx0gEIRquoQH1P7yIEJ3xOGoxQfIRVpls=
+ <YOUR_DEX_CA_CERTIFICATE_BASE64_HERE>
-----END CERTIFICATE-----🤖 Prompt for AI Agents
In `@docs/en/installation/kubeflow.mdx` around lines 33 - 51, The PEM block
beginning with "-----BEGIN CERTIFICATE-----" and ending with "-----END
CERTIFICATE-----" contains a real CA certificate; replace that entire PEM block
with a clearly labeled placeholder (e.g., "-----BEGIN
CERTIFICATE-----\n<PLACEHOLDER CERTIFICATE FOR DOCUMENTATION ONLY>\n-----END
CERTIFICATE-----") and add a short inline note that this is a dummy/example
certificate for documentation purposes only so no real PKI material is
published.
|
|
||
| ## required for kubeflow pipelines | ||
| ## TEMPLATE: "cluster.local/ns/<KUBEFLOW_NAMESPACE>/sa/<KFP_UI_SERVICE_ACCOUNT>" | ||
| - "cluster.local/ns/kubeflow-new/sa/ml-pipeline-ui" |
There was a problem hiding this comment.
kubeflow-new looks like an environment-specific namespace — should this be a placeholder?
The principal on line 176 references kubeflow-new as the namespace (cluster.local/ns/kubeflow-new/sa/ml-pipeline-ui), but the rest of the document doesn't mention this namespace. If this is meant to be the Kubeflow installation namespace, consider using a placeholder like <kubeflow-namespace> or documenting what kubeflow-new refers to, so users configure it correctly for their environment.
🤖 Prompt for AI Agents
In `@docs/en/installation/kubeflow.mdx` at line 176, Replace the hard-coded
namespace string "cluster.local/ns/kubeflow-new/sa/ml-pipeline-ui" with a clear
placeholder (for example
"cluster.local/ns/<kubeflow-namespace>/sa/ml-pipeline-ui") or add a brief
clarifying sentence near that occurrence explaining what "kubeflow-new" refers
to (i.e., the Kubeflow installation namespace) so readers know to substitute
their environment-specific namespace; update the text around the symbol to use
the placeholder or the explanatory note consistently.
Deploying alauda-ai with
|
| Latest commit: |
e7a0ac6
|
| Status: | ✅ Deploy successful! |
| Preview URL: | https://32f4f4cd.alauda-ai.pages.dev |
| Branch Preview URL: | https://add-kubeflow-deploy-doc.alauda-ai.pages.dev |
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Fix all issues with AI agents
In `@docs/en/installation/kubeflow.mdx`:
- Line 19: Capitalize the first word of the sentence "obtain the platform dex CA
certificate for later use:" so it reads "Obtain the platform dex CA certificate
for later use:"; update this sentence in docs/en/installation/kubeflow.mdx to
start with an uppercase "O".
🧹 Nitpick comments (5)
docs/en/installation/kubeflow.mdx (5)
85-91: Consider using a placeholder for the IP address.
192.168.171.123appears to be an environment-specific address. Using a placeholder like<platform-address>would be more consistent with the<platform_password>placeholder already used on the same line, and clearer for readers who need to substitute their own values.📝 Proposed fix
-violet push --platform-address="https://192.168.171.123" \ +violet push --platform-address="https://<platform_address>" \
109-114: Replace hardcoded IP in redirect URI with a placeholder.
192.168.139.133:30665is environment-specific. Use a placeholder so readers know to substitute their own value, consistent with the comment on line 112.📝 Proposed fix
redirectURIs: - ... # Add the following line, note: the redirect address must be consistent with oidcRedirectURL in values: - - https://192.168.139.133:30665/* + - https://<your-kubeflow-host>:<port>/*
127-136: Addyamllanguage specifier to the code block.This code block contains a Kubernetes YAML manifest but lacks a language identifier, which will result in no syntax highlighting. The same issue applies to the code block at line 142.
📝 Proposed fix
-``` +```yaml apiVersion: kubeflow.org/v1beta1 kind: Profile
142-186: Addyamllanguage specifier to the code block.Same as the earlier block — this YAML manifest should have the
yamllanguage identifier for proper syntax highlighting.📝 Proposed fix
-``` +```yaml apiVersion: rbac.authorization.k8s.io/v1 kind: RoleBinding
138-138: Rephrase the heading to be more concise.Using a conditional sentence as a section heading is unusual and verbose. Consider a shorter, scannable heading.
📝 Proposed fix
-### 3. If binding user to an already created namespace, you also need to complete the following configuration: +### 3. Bind User to an Existing Namespace
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Fix all issues with AI agents
In `@docs/en/installation/kubeflow.mdx`:
- Line 70: The X-Auth-Request-Redirect header hardcodes the http scheme which
can downgrade HTTPS deployments; update the header value for
X-Auth-Request-Redirect to use https://%REQ(Host)%%REQ(:PATH)% (or a
protocol-agnostic/existing proxy variable if available) and add a short note in
the surrounding text explaining when to use https vs http and that the scheme
must match the platform's public URL; locate the header line containing
"X-Auth-Request-Redirect" and adjust the value accordingly and update the nearby
examples that assume HTTPS to remain consistent.
🧹 Nitpick comments (6)
docs/en/installation/kubeflow.mdx (6)
5-8: Inconsistent capitalization of "Kubeflow" in plugin descriptions.Line 5 capitalizes "Kubeflow base components" while lines 6–8 use lowercase "kubeflow". Consider capitalizing consistently for a polished look.
📝 Proposed fix
-- kfmr: kubeflow model registry instance (Helm Chart) -- kfp: kubeflow pipeline -- kftraining: kubeflow training operator (deprecated) +- kfmr: Kubeflow model registry instance (Helm Chart) +- kfp: Kubeflow pipeline +- kftraining: Kubeflow training operator (deprecated)
85-91: Use placeholders instead of hardcoded IP/email in the example command.Lines 87–88 use a specific IP address and email that differ from other examples in the doc (e.g., line 113 uses a different IP). Despite the comment on line 86, using explicit placeholders would be clearer and avoid users accidentally copying the wrong address.
📝 Proposed fix
# Note: replace your platform addr, username, password and cluster name. -violet push --platform-address="https://192.168.171.123" \ - --platform-username="admin@cpaas.io" \ +violet push --platform-address="https://<PLATFORM_ADDRESS>" \ + --platform-username="<PLATFORM_USERNAME>" \ --platform-password="<platform_password>" \ <your downloaded plugin package file>
109-114: Use a placeholder for the environment-specific redirect URI.The redirect URI on line 113 uses a hardcoded IP and port. Replace with a placeholder so users don't accidentally copy the wrong address.
📝 Proposed fix
redirectURIs: - ... # Add the following line, note: the redirect address must be consistent with oidcRedirectURL in values: - - https://192.168.139.133:30665/* + - https://<KUBEFLOW_HOST>:<PORT>/*
127-136: Addyamllanguage identifier to the fenced code block for syntax highlighting.The code block starting at line 127 lacks a language identifier. This also applies to the code block at line 142.
📝 Proposed fix (line 127)
-``` +```yaml apiVersion: kubeflow.org/v1beta1
138-138: Heading is overly long and reads like a sentence — consider shortening.Section headings should be concise. Consider rephrasing:
-### 3. If binding user to an already created namespace, you also need to complete the following configuration: +### 3. Bind a User to an Existing NamespaceThen move the explanatory text into the section body.
202-207: Capitalize "MySQL" as a proper product name.Lines 203–205 use lowercase "mysql" in parameter descriptions. "MySQL" is a proper product name and should be capitalized consistently.
📝 Proposed fix
-- mysqlStorageClass: The mysql storage class used by Model Registry. Needs to be a storage class supported by the target deployment cluster. -- mysqlStorageSize: The mysql storage size used by Model Registry. -- mysqlDataBase: Database name (will be created automatically). +- mysqlStorageClass: The MySQL storage class used by Model Registry. Needs to be a storage class supported by the target deployment cluster. +- mysqlStorageSize: The MySQL storage size used by Model Registry. +- mysqlDataBase: The database name (will be created automatically).
docs/en/installation/kubeflow.mdx
Outdated
| - x-auth-request-email | ||
| - x-auth-request-access-token | ||
| includeAdditionalHeadersInCheck: | ||
| X-Auth-Request-Redirect: http://%REQ(Host)%%REQ(:PATH)% |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
cat -n docs/en/installation/kubeflow.mdx | sed -n '50,130p'Repository: alauda/aml-docs
Length of output: 4012
Redirect scheme is hardcoded to http:// — should this be https://?
The X-Auth-Request-Redirect header uses http:// but the rest of the document uses HTTPS URLs (lines 87, 113). If the platform is served over HTTPS, this would redirect the user to an HTTP URL after authentication, potentially breaking the OAuth flow or causing a security downgrade.
Consider using https:// or, if the scheme varies, documenting when each should be used:
Suggested diff
- X-Auth-Request-Redirect: http://%REQ(Host)%%REQ(:PATH)%
+ X-Auth-Request-Redirect: https://%REQ(Host)%%REQ(:PATH)%🤖 Prompt for AI Agents
In `@docs/en/installation/kubeflow.mdx` at line 70, The X-Auth-Request-Redirect
header hardcodes the http scheme which can downgrade HTTPS deployments; update
the header value for X-Auth-Request-Redirect to use
https://%REQ(Host)%%REQ(:PATH)% (or a protocol-agnostic/existing proxy variable
if available) and add a short note in the surrounding text explaining when to
use https vs http and that the scheme must match the platform's public URL;
locate the header line containing "X-Auth-Request-Redirect" and adjust the value
accordingly and update the nearby examples that assume HTTPS to remain
consistent.
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Fix all issues with AI agents
In `@docs/en/installation/kubeflow.mdx`:
- Around line 94-98: Update the intro paragraph to either list the "volcano"
plugin alongside the other plugins or explicitly mark it as optional;
specifically reference the "volcano" plugin and clarify it is only required when
enabling volcano scheduler support for "kftraining" so readers know it's not
mandatory for all Kubeflow onboardings. Ensure the introduction text that
enumerates plugins includes either the volcano entry or a parenthetical note
linking volcano to optional kftraining scheduler support.
- Around line 110-121: Update the documentation block that instructs editing the
ConfigMap under cpaas-system: specify the exact ConfigMap name (e.g., "dex" or
"dex-config") so users can find it, remove the duplicate "note" so the line
reads only "Note:", and replace the hardcoded host/port in the redirectURIs
example (currently "192.168.139.133:30665") with a placeholder like
"<KUBEFLOW_HOST>:<PORT>", while keeping the reference to oidcRedirectURL and the
redirectURIs key unchanged.
🧹 Nitpick comments (2)
docs/en/installation/kubeflow.mdx (2)
134-143: Addyamllanguage identifier to the code block.This code block contains a Kubernetes YAML manifest but lacks a language identifier. Adding
yamlenables syntax highlighting and improves readability in the rendered documentation.-``` +```yaml apiVersion: kubeflow.org/v1beta1 kind: ProfileThe same applies to the code block starting at line 149.
145-147: Section heading reads as a conditional clause rather than a title.
"3. If binding user to an already created namespace, you also need to complete the following configuration:"is awkward as a heading. Consider rephrasing for clarity:-### 3. If binding user to an already created namespace, you also need to complete the following configuration: +### 3. Bind a User to an Existing NamespaceThe conditional context can be explained in the body text instead.
docs/en/installation/kubeflow.mdx
Outdated
| - kfbase: Kubeflow Base functionality | ||
| - chart-kubeflow-model-registry: Kubeflow Model Registry | ||
| - kfp: Kubeflow Pipeline functionality | ||
| - kftraining: Kubeflow Training Operator (deprecated) | ||
| - volcano: volcano scheduler |
There was a problem hiding this comment.
volcano plugin not mentioned in the introduction.
Line 98 lists volcano: volcano scheduler as a plugin to onboard, but the introduction (lines 5–8) doesn't mention it. Either add it to the introduction or clarify that it's an optional dependency (only needed when enabling volcano scheduler support for kftraining, per line 100).
🤖 Prompt for AI Agents
In `@docs/en/installation/kubeflow.mdx` around lines 94 - 98, Update the intro
paragraph to either list the "volcano" plugin alongside the other plugins or
explicitly mark it as optional; specifically reference the "volcano" plugin and
clarify it is only required when enabling volcano scheduler support for
"kftraining" so readers know it's not mandatory for all Kubeflow onboardings.
Ensure the introduction text that enumerates plugins includes either the volcano
entry or a parenthetical note linking volcano to optional kftraining scheduler
support.
| In **Administrator** - **Clusters** - **Resources**, select `Global` cluster, | ||
| find the `ConfigMap` resource in the `cpaas-system` namespace, and click the | ||
| edit button to add the following configuration under `redirectURIs`: | ||
|
|
||
| > **Note: note: the redirect host and port must be the same with `oidcRedirectURL` configured when installing the "Kubeflow Base" plugin.** | ||
|
|
||
| ```yaml | ||
| redirectURIs: | ||
| - ... | ||
| # Add the following line, | ||
| - https://192.168.139.133:30665/* | ||
| ``` |
There was a problem hiding this comment.
Missing ConfigMap name, hardcoded IP, and duplicate "note" in this section.
Three issues in this block:
- Line 111: Which
ConfigMap? There are likely many incpaas-system. Specify the resource name (e.g.,dexordex-config) so users can locate it. - Line 114:
"Note: note:"— double "note" with extra whitespace is a typo. Should be just"Note:". - Line 120: The IP
192.168.139.133:30665is hardcoded. Use a placeholder like<KUBEFLOW_HOST>:<PORT>to avoid confusion with the other hardcoded IP on line 88.
Suggested fixes
-> **Note: note: the redirect host and port must be the same with `oidcRedirectURL` configured when installing the "Kubeflow Base" plugin.**
+> **Note: The redirect host and port must match the `oidcRedirectURL` configured when installing the "Kubeflow Base" plugin.**- - https://192.168.139.133:30665/*
+ - https://<KUBEFLOW_HOST>:<PORT>/*🤖 Prompt for AI Agents
In `@docs/en/installation/kubeflow.mdx` around lines 110 - 121, Update the
documentation block that instructs editing the ConfigMap under cpaas-system:
specify the exact ConfigMap name (e.g., "dex" or "dex-config") so users can find
it, remove the duplicate "note" so the line reads only "Note:", and replace the
hardcoded host/port in the redirectURIs example (currently
"192.168.139.133:30665") with a placeholder like "<KUBEFLOW_HOST>:<PORT>", while
keeping the reference to oidcRedirectURL and the redirectURIs key unchanged.
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Fix all issues with AI agents
In `@docs/en/installation/kubeflow.mdx`:
- Line 91: Replace the hardcoded IP in the example command using the same
placeholder used elsewhere: change the argument to use a placeholder (e.g.,
<PLATFORM_ADDRESS>) in the "violet push --platform-address" example so it reads
"violet push --platform-address=\"<PLATFORM_ADDRESS>\" ..." ensuring it matches
the placeholder referenced on the previous line and avoids embedding
environment-specific IPs.
🧹 Nitpick comments (3)
docs/en/installation/kubeflow.mdx (3)
137-146: YAML code blocks missing language identifier.These YAML blocks use bare
```fences without a language hint (e.g.,```yaml). This will result in no syntax highlighting in the rendered documentation.Suggested fix (apply to both blocks)
-``` +```yaml apiVersion: kubeflow.org/v1beta1-``` +```yaml apiVersion: rbac.authorization.k8s.io/v1Also applies to: 152-196
1-10: Introduction doesn't mentionvolcanoorLWSas optional/required dependencies.Line 103 mentions volcano as needed for kftraining scheduler support, and line 17/230 mentions LWS as a dependency of kubeflow-trainer. Consider listing these in the introduction or adding a "Dependencies" subsection so users have a complete picture upfront.
148-148: Section heading style: numbered heading reads awkwardly as a conditional instruction.
### 3. If binding user to an already created namespace, you also need to complete the following configuration:is unusually long and conditional for a heading. Consider shortening it, e.g.,### 3. Bind User to an Existing Namespace.
|
|
||
| ```bash | ||
| # Note: replace your platform addr, username, password and cluster name. | ||
| violet push --platform-address="https://192.168.171.123" \ |
There was a problem hiding this comment.
Hardcoded IP address in the example command.
192.168.171.123 is a specific environment IP. Consider using a placeholder like <PLATFORM_ADDRESS> consistent with the instruction on line 90.
Suggested fix
-violet push --platform-address="https://192.168.171.123" \
+violet push --platform-address="https://<PLATFORM_ADDRESS>" \🤖 Prompt for AI Agents
In `@docs/en/installation/kubeflow.mdx` at line 91, Replace the hardcoded IP in
the example command using the same placeholder used elsewhere: change the
argument to use a placeholder (e.g., <PLATFORM_ADDRESS>) in the "violet push
--platform-address" example so it reads "violet push
--platform-address=\"<PLATFORM_ADDRESS>\" ..." ensuring it matches the
placeholder referenced on the previous line and avoids embedding
environment-specific IPs.
1 participant
Summary by CodeRabbit