docs(self-managed): promote 0.6.1 as latest stable - #605
Conversation
Signed-off-by: Stephanie Baum <sbaum@nvidia.com>
📝 WalkthroughWalkthroughThe PR adds the self-managed NVCF 0.6.1 documentation set. It covers installation, cluster management, APIs, workloads, observability, operations, upgrades, and troubleshooting. It also promotes 0.6.1 from release candidate to the latest stable documentation version. ChangesSelf-managed NVCF documentation
Estimated code review effort: 5 (Critical) | ~120 minutes Possibly related PRs
Suggested labels: Suggested reviewers: 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
|
🌿 Preview your docs: https://nvidia-preview-docs-promote-0-6-1.docs.buildwithfern.com/nvcf |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
Note
Due to the large number of review comments, Critical severity comments were prioritized as inline comments.
🟡 Minor comments (44)
docs/v0.6.1/fake-gpu-operator.md-65-67 (1)
65-67: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winSpecify a language for the error example.
Use
texton the opening fence. The current untyped fence fails MD040.Run the documentation check after this correction. As per coding guidelines, "Run
./tools/ci/check-docsbefore completing documentation changes."🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/fake-gpu-operator.md` around lines 65 - 67, Update the error example code fence in fake-gpu-operator documentation to declare the text language, resolving the MD040 documentation check; then run ./tools/ci/check-docs to verify the documentation change.Sources: Coding guidelines, Linters/SAST tools
docs/v0.6.1/cluster-management/configuration.md-302-369 (1)
302-369: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winFix the nested fenced code blocks.
The unindented closing fences cause the docs checker to treat Lines 352 and 366 as new fenced blocks without a language. Indent each closing fence to match its code block inside the ordered list.
Run the documentation check after this correction. As per coding guidelines, "Run
./tools/ci/check-docsbefore completing documentation changes."🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/cluster-management/configuration.md` around lines 302 - 369, Fix the nested fenced code blocks in the ordered-list steps by indenting each closing fence to match its corresponding opening fence and list indentation, including the YAML, kubectl apply, and kubectl logs blocks. Preserve the existing commands and content, then run ./tools/ci/check-docs to verify the documentation.Sources: Coding guidelines, Linters/SAST tools
docs/v0.6.1/cluster-management/kai-scheduler.md-27-73 (1)
27-73: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winClose the YAML code block correctly.
The docs checker treats the fence at Line 72 as a new code block without a language. Use the MDX fence structure that correctly closes the YAML block inside
<Accordion>.Run the documentation check after this correction. As per coding guidelines, "Run
./tools/ci/check-docsbefore completing documentation changes."🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/cluster-management/kai-scheduler.md` around lines 27 - 73, Correct the Markdown/MDX fence structure in the Accordion containing “kai-scheduler-queues.yaml” so the YAML code block is properly closed rather than starting an untyped block. Preserve the YAML content and run ./tools/ci/check-docs to verify the documentation.Sources: Coding guidelines, Linters/SAST tools
docs/v0.6.1/cluster-management/configuration.md-18-18 (1)
18-18: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winCorrect the node selector description.
Replace
specicwithspecific.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/cluster-management/configuration.md` at line 18, In the node selector description table entry, correct the typo “specic” to “specific” while leaving the rest of the documentation unchanged.Source: Linters/SAST tools
docs/v0.6.1/fake-gpu-operator.md-141-141 (1)
141-141: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winRemove the obsolete
user/path segment.
docs/v0.6.1/fake-gpu-operator.md#L141-L141: change the control-plane installation link to./helmfile-installation.md.docs/v0.6.1/fake-gpu-operator.md#L163-L164: change the self-managed cluster link to./cluster-management/self-managed.md.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/fake-gpu-operator.md` at line 141, Update both links in docs/v0.6.1/fake-gpu-operator.md: change the control-plane installation link at lines 141-141 to ./helmfile-installation.md, and change the self-managed cluster link at lines 163-164 to ./cluster-management/self-managed.md.docs/v0.6.1/cluster-management/configuration.md-32-35 (1)
32-35: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winRemove prohibited Markdown emphasis and non-ASCII punctuation.
docs/v0.6.1/cluster-management/configuration.md#L32-L35: replace Markdown bold with plain text or headings.docs/v0.6.1/cluster-management/container-cache.md#L13-L15: replace Markdown bold with plain text or headings.docs/v0.6.1/cluster-management/gxcache.md#L13-L15: replace Markdown bold with plain text or headings.docs/v0.6.1/cluster-management/kai-scheduler.md#L7-L10: replace Markdown bold with plain text or headings.docs/v0.6.1/cluster-management/reference.md#L17-L25: replace Markdown bold and the em dash with ASCII text.docs/v0.6.1/fake-gpu-operator.md#L7-L10: replace Markdown bold with plain text or headings.docs/v0.6.1/cluster-management/monitoring.md#L10-L12: replace Markdown bold with plain text or headings.As per coding guidelines, "Documentation must use no Markdown bold for emphasis, no emojis, no em dash, succinct direct sentences, and only standard ASCII in committed text."
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/cluster-management/configuration.md` around lines 32 - 35, Remove Markdown bold emphasis from docs/v0.6.1/cluster-management/configuration.md lines 32-35, docs/v0.6.1/cluster-management/container-cache.md lines 13-15, docs/v0.6.1/cluster-management/gxcache.md lines 13-15, docs/v0.6.1/cluster-management/kai-scheduler.md lines 7-10, docs/v0.6.1/fake-gpu-operator.md lines 7-10, and docs/v0.6.1/cluster-management/monitoring.md lines 10-12, replacing it with plain text or headings. In docs/v0.6.1/cluster-management/reference.md lines 17-25, remove Markdown bold and replace the em dash with ASCII text; preserve the existing documentation meaning and use only standard ASCII.Source: Coding guidelines
docs/v0.6.1/samples/scripts/create-nvcr-pull-secrets.sh-7-7 (1)
7-7: 🔒 Security & Privacy | 🟡 Minor | ⚡ Quick winRequire an injected
NGC_API_KEY.Do not instruct users to place an NGC API key in the script. Fail before creating secrets when the environment variable is unset.
Proposed fix
-export NGC_API_KEY="<your-ngc-api-key>" +: "${NGC_API_KEY:?Set NGC_API_KEY before running this script}"🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/samples/scripts/create-nvcr-pull-secrets.sh` at line 7, Update the script’s NGC_API_KEY handling to require the value from the existing environment rather than assigning a placeholder or embedded key. Add an early validation that exits before any secret creation when NGC_API_KEY is unset, while preserving the current secret-generation flow when it is provided.Source: Linters/SAST tools
docs/v0.6.1/local-development.md-8-10 (1)
8-10: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winRemove Markdown bold from the documentation.
The documentation guidelines prohibit Markdown bold in
*.mdfiles. Use plain text or code formatting forlocal development only,Single-cluster,Multi-cluster,CLI, andHelmfile.Also applies to: 33-41, 56-68
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/local-development.md` around lines 8 - 10, Remove Markdown bold formatting from the specified documentation terms in local-development.md, including “local development only,” “Single-cluster,” “Multi-cluster,” “CLI,” and “Helmfile,” while preserving their plain text content and surrounding documentation.Source: Coding guidelines
docs/v0.6.1/quickstart.md-87-101 (1)
87-101: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winList Bazel as a prerequisite for the repository build path.
The documented
bazel buildcommand cannot run unless Bazel or Bazelisk is installed. Add it to the prerequisites, or make the packaged CLI path the default and mark the Bazel path as optional.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/quickstart.md` around lines 87 - 101, Update the prerequisites in the quickstart section for building nvcf-cli so they explicitly require Bazel or Bazelisk before the documented bazel build commands, or clearly label this repository build path as optional and provide the packaged CLI path as the default.docs/v0.6.1/csp-end-to-end-example-installation.md-42-50 (1)
42-50: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winSpecify a language for the numbered code block.
markdownlintreports this fence because it has no language. Usetextfor the installation-order block.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/csp-end-to-end-example-installation.md` around lines 42 - 50, Specify the text language on the numbered installation-order fenced block in the end-to-end installation documentation, changing only the fence annotation while preserving the block contents.Source: Linters/SAST tools
docs/v0.6.1/helmfile-installation.md-194-194 (1)
194-194: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winUse the
0.6.1sample files.This
0.6.1guide links to rawdocs/v0.6.0examples. The repository provides the corresponding0.6.1samples. Link to the local files so the guide remains aligned with this release.Also applies to: 589-591
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/helmfile-installation.md` at line 194, Update the EKS sample references in the v0.6.1 Helmfile installation guide to use the corresponding local 0.6.1 sample files instead of raw docs/v0.6.0 URLs, including the related references around the additional affected lines.docs/v0.6.1/local-development.md-74-88 (1)
74-88: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winComplete the prerequisites for the documented commands.
The local-development pages build
nvcf-cliwith Go and parse verification responses with Python 3, but the prerequisite lists omit those tools.
docs/v0.6.1/local-development.md#L74-L88: add Go to the common prerequisites.docs/v0.6.1/local-development/multi-cluster-cli.md#L40-L59: add Go.docs/v0.6.1/local-development/multi-cluster-cli.md#L308-L316: add Python 3.docs/v0.6.1/local-development/multi-cluster-helmfile.md#L40-L60: add Go.docs/v0.6.1/local-development/multi-cluster-helmfile.md#L250-L257: add Python 3.docs/v0.6.1/local-development/single-cluster-cli.md#L20-L39: add Go.docs/v0.6.1/local-development/single-cluster-cli.md#L266-L273: add Python 3.docs/v0.6.1/local-development/single-cluster-helmfile.md#L20-L40: add Go.docs/v0.6.1/local-development/single-cluster-helmfile.md#L217-L224: add Python 3.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/local-development.md` around lines 74 - 88, Complete the prerequisite lists for the documented commands: add Go to docs/v0.6.1/local-development.md (74-88), docs/v0.6.1/local-development/multi-cluster-cli.md (40-59), docs/v0.6.1/local-development/multi-cluster-helmfile.md (40-60), docs/v0.6.1/local-development/single-cluster-cli.md (20-39), and docs/v0.6.1/local-development/single-cluster-helmfile.md (20-40); add Python 3 to docs/v0.6.1/local-development/multi-cluster-cli.md (308-316), docs/v0.6.1/local-development/multi-cluster-helmfile.md (250-257), docs/v0.6.1/local-development/single-cluster-cli.md (266-273), and docs/v0.6.1/local-development/single-cluster-helmfile.md (217-224).docs/v0.6.1/example-dashboards.md-164-173 (1)
164-173: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winHyphenate the compound phrase.
Change
chicken and egg problemtochicken-and-egg problem.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/example-dashboards.md` around lines 164 - 173, Update the compound phrase in the Step 2 documentation text to use “chicken-and-egg problem” instead of “chicken and egg problem.”Source: Linters/SAST tools
docs/v0.6.1/observability.md-196-196 (1)
196-196: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winHyphenate the compound modifier.
Change
5 minute windowto5-minute window.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/observability.md` at line 196, Update the latency description to hyphenate the compound modifier, changing “5 minute window” to “5-minute window” while preserving the rest of the text.Source: Linters/SAST tools
docs/v0.6.1/metrics/state-metrics/metrics.md-10-12 (1)
10-12: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winRemove the duplicate
nvcf_function_instances_currententry.The same metric and label set appears twice with different descriptions. Keep one canonical definition, or use different metric names if the semantics differ.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/state-metrics/metrics.md` around lines 10 - 12, Remove the duplicate nvcf_function_instances_current row from the metrics table, retaining one canonical definition with its intended description and labels; only introduce a different metric name if the two entries represent distinct semantics.docs/v0.6.1/observability.md-27-32 (1)
27-32: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winUse plain labels or real links in the capability list.
[Metrics Collection]:,[Logging]:,[Tracing]:, and[Dashboards]:have no destinations. In list items, they render as bracketed text rather than links. Remove the brackets or add actual destinations.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/observability.md` around lines 27 - 32, Update the capability list in the observability documentation to remove the unused link-style brackets from Metrics Collection, Logging, Tracing, and Dashboards, unless valid destinations are available; preserve the existing descriptions and list structure.docs/v0.6.1/grpc-load-test-sli-guide.md-50-65 (1)
50-65: 🗄️ Data Integrity & Integration | 🟡 Minor | ⚡ Quick winUse the documented session-init histogram name.
Change the heading to
nvcf_grpc_proxy_service_session_init_secondsand querynvcf_grpc_proxy_service_session_init_seconds_bucket. The current_total_bucketseries is not documented.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/grpc-load-test-sli-guide.md` around lines 50 - 65, Update the session-init metric references in the section for nvcf_grpc_proxy_service_session_init_seconds_total (p95): change the heading to nvcf_grpc_proxy_service_session_init_seconds and use nvcf_grpc_proxy_service_session_init_seconds_bucket in the histogram_quantile query, preserving the existing labels and rate window.docs/v0.6.1/grpc-load-testing.md-7-10 (1)
7-10: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winUse the versioned CLI document for the versioned guide. Change both
../user/cli.mdlinks to./cli.md. Fern maps the v0.6.1 CLI page todocs/v0.6.1/cli.md.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/grpc-load-testing.md` around lines 7 - 10, Update both links in the introductory nvcf-cli setup instructions to use the version-local target ./cli.md instead of ../user/cli.md, preserving their existing link text and surrounding guidance.docs/v0.6.1/observability.md-394-396 (1)
394-396: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winReplace the unrelated redaction reference.
The link targets .NET log redaction documentation, not OpenTelemetry Collector configuration. Link to the Collector Transform or Redaction processor documentation.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/observability.md` around lines 394 - 396, Update the observability documentation’s log redaction reference to point to OpenTelemetry Collector Transform or Redaction processor documentation instead of the unrelated .NET implementation link. Keep the surrounding guidance about collector responsibility and sensitive data redaction unchanged.docs/v0.6.1/metrics/utils-container/metrics.md-8-8 (1)
8-8: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winComplete the
nvcf_worker_service_response_totalrow.Set the type to
Counter, adderror_codeas an interesting label, and describe the total number of service responses processed, including error codes.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/utils-container/metrics.md` at line 8, Update the nvcf_worker_service_response_total row in the metrics documentation: change its type to Counter, add error_code as the interesting label, and describe it as the total number of processed service responses including error codes.docs/v0.6.1/http-soak-testing.md-341-341 (1)
341-341: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winReplace the non-ASCII ellipsis.
The
FUNCTIONSexample contains…. Use an ASCII placeholder such as<id>.As per coding guidelines, documentation must use only standard ASCII in committed text.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/http-soak-testing.md` at line 341, Replace the non-ASCII ellipsis in the FUNCTIONS JSON example with an ASCII placeholder such as <id>, preserving the documented array and funcId structure.Source: Coding guidelines
docs/v0.6.1/http-soak-testing.md-26-31 (1)
26-31: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winRemove Markdown bold from the documentation.
**load_tester_supreme**and**Default control plane sizing:**violate the documentation formatting rule. Use plain text or inline code instead.As per coding guidelines, documentation must use no Markdown bold for emphasis, no emojis, no em dash, succinct direct sentences, and only standard ASCII in committed text.
Also applies to: 313-320
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/http-soak-testing.md` around lines 26 - 31, Remove Markdown bold formatting from the documentation, including load_tester_supreme and Default control plane sizing, while preserving their text and meaning. Use plain text or inline code where code-style emphasis is needed, and ensure the affected sections contain no bold markers or other non-ASCII emphasis characters.Source: Coding guidelines
docs/v0.6.1/metrics/grpc-proxy/metrics.md-27-28 (1)
27-28: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winRemove non-ASCII and bold markup.
**gRPC**violates the no-bold rule.→is not standard ASCII. UsegRPCandgRPC proxy -> NATS pipe.As per coding guidelines, documentation must use no Markdown bold for emphasis, no emojis, no em dash, succinct direct sentences, and only standard ASCII in committed text.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/grpc-proxy/metrics.md` around lines 27 - 28, Update the affected metrics documentation text to remove bold Markdown around gRPC and replace the non-ASCII arrow in “gRPC proxy → NATS pipe” with the ASCII sequence “->”. Preserve the existing wording and meaning while ensuring the committed text uses only standard ASCII.Source: Coding guidelines
docs/v0.6.1/metrics/metrics-index.md-3-3 (1)
3-3: 🗄️ Data Integrity & Integration | 🟡 Minor | ⚡ Quick winAlign the overview claim with the linked pages.
The overview says every linked page lists metric names, types, sources, descriptions, labels, and filters.
docs/v0.6.1/metrics/llm-function-invocation-path.mdlists only metrics and labels. The Vault/OpenBao entry also points to upstream documentation instead of listing the metadata. Change the claim or add the missing metadata.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/metrics-index.md` at line 3, Update the overview sentence in the metrics index so it accurately describes the linked pages, removing metadata claims that are not consistently provided by pages such as llm-function-invocation-path and the Vault/OpenBao entry.docs/v0.6.1/metrics/invocation-service/metrics.md-12-12 (1)
12-12: 🗄️ Data Integrity & Integration | 🟡 Minor | ⚡ Quick winCorrect the
nats_jetstream_publishdescription.The metric name identifies publish operations, but the description says “NATS Stream creation count”. These are different events. Describe the event that the metric actually counts.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/invocation-service/metrics.md` at line 12, Update the description column for the nats_jetstream_publish metric in the metrics table to describe the count of NATS JetStream publish operations, replacing the incorrect stream-creation wording while leaving the other fields unchanged.docs/v0.6.1/metrics/grpc-proxy/metrics.md-32-37 (1)
32-37: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winMake the reproduction command cover the documented metric families.
The command filters only
nvcf_grpc_proxy_*. It does not display the documentedhttp_server_*orrpc_client_*metrics. Remove the filter or include all documented prefixes.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/grpc-proxy/metrics.md` around lines 32 - 37, Update the reproduction curl command in the “Reproducing locally” section so its metrics output includes the documented nvcf_grpc_proxy, http_server, and rpc_client metric families; remove the restrictive grep or expand its prefixes accordingly.docs/v0.6.1/metrics/metrics-index.md-10-10 (1)
10-10: 🗄️ Data Integrity & Integration | 🟡 Minor | ⚡ Quick winAdd gRPC Proxy to Fern navigation. The page exists, but
fern/versions/v0.6.1.ymlhas no corresponding entry.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/metrics-index.md` at line 10, Add the existing gRPC Proxy metrics page to the Fern navigation configuration in the v0.6.1 version manifest, using the established navigation entry format and linking to the metrics page without changing the page content.docs/v0.6.1/http-soak-testing.md-234-237 (1)
234-237: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winAdd the
invoke_latency_msthreshold.The script records
invoke_latency_ms, and the results table requires its p95 to stay below 5,000 ms. Addinvoke_latency_ms: ['p(95)<5000']tooptions.thresholds; otherwise, mark this metric as informational.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/http-soak-testing.md` around lines 234 - 237, Update the options.thresholds configuration to include invoke_latency_ms with a p95 limit below 5,000 ms, matching the results table requirement. If this metric is intentionally informational instead, remove it from the required threshold expectations and document that behavior consistently.docs/v0.6.1/metrics/invocation-service/metrics.md-5-8 (1)
5-8: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winCorrect the metrics table entries.
- Fix the malformed
namespacematcher tonamespace="astro-tenant-nvcf-invocation-service".- Keep the
exported_endpointmatcher. Its backslashes are Markdown escapes and render as^/v2/nvcf.*|^/health$.- Change the
nats_jetstream_publishdescription toNumber of messages published to NATS streams.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/invocation-service/metrics.md` around lines 5 - 8, Correct the metrics table entries by changing the malformed namespace matcher on app_invocation_error to namespace="astro-tenant-nvcf-invocation-service", while preserving the existing exported_endpoint matcher and its Markdown escapes. Update the nats_jetstream_publish description to “Number of messages published to NATS streams.”docs/v0.6.1/streaming-functions.md-108-110 (1)
108-110: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winCorrect the WebSocket sentence.
Change
to accepting WebSocket connectiontoto accept WebSocket connections.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/streaming-functions.md` around lines 108 - 110, Correct the WebSocket server description in the streaming functions documentation by changing the wording to “to accept WebSocket connections.”Source: Linters/SAST tools
docs/v0.6.1/lls-installation.md-18-20 (1)
18-20: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winApply the documentation formatting rule consistently.
The affected files use Markdown bold for emphasis.
docs/v0.6.1/lls-installation.mdalso uses an em dash. Replace these with plain ASCII text.
docs/v0.6.1/lls-installation.md#L18-L20: remove Markdown bold emphasis.docs/v0.6.1/lls-installation.md#L401-L404: remove Markdown bold and replace the em dash with ASCII punctuation.docs/v0.6.1/streaming-functions.md#L71-L75: remove Markdown bold emphasis.docs/v0.6.1/third-party-registries.md#L7-L10: remove Markdown bold emphasis.As per coding guidelines, documentation must use no Markdown bold for emphasis, no emojis, no em dash, succinct direct sentences, and only standard ASCII in committed text.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/lls-installation.md` around lines 18 - 20, Remove Markdown bold emphasis from docs/v0.6.1/lls-installation.md lines 18-20 and 401-404, replacing the em dash at lines 401-404 with standard ASCII punctuation; also remove Markdown bold emphasis from docs/v0.6.1/streaming-functions.md lines 71-75 and docs/v0.6.1/third-party-registries.md lines 7-10. Preserve the wording while ensuring all affected documentation uses plain ASCII text without bold or em dashes.Source: Coding guidelines
docs/v0.6.1/lls-installation.md-67-68 (1)
67-68: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winCorrect the spelling errors.
Use
Streaminginstead ofStramingandInternalinstead ofInteral.Also applies to: 647-650
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/lls-installation.md` around lines 67 - 68, Correct the spelling errors in the documentation legend: update “Straming” to “Streaming” and “Interal” to “Internal” in the affected descriptions, including the additional occurrence referenced by the comment.Source: Linters/SAST tools
docs/v0.6.1/lls-installation.md-397-402 (1)
397-402: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winCorrect the step references.
The text says to save
TG_ARNfor Step 3, but the value is used in Step 4.2. The next heading jumps from Step 4 to Step 6. Correct the reference and renumber or add Step 5 so the procedure has one clear sequence.Also applies to: 511-513
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/lls-installation.md` around lines 397 - 402, Update the installation procedure’s step references around the TG_ARN instructions: identify the correct step where TG_ARN is consumed (Step 4.2), change the save note to reference that step, and fix the later heading sequence so it progresses continuously from Step 4 through Step 6 by renumbering or adding the missing Step 5.docs/v0.6.1/autoscaling/operations.md-59-59 (1)
59-59: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winFix the broken architecture anchor.
./architecture.md#cassandra-lightweight-transactions-lwtsdoes not match any heading indocs/v0.6.1/autoscaling/architecture.md. Link to the existingCoordination and Self-Healingheading or add the referenced heading.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/autoscaling/operations.md` at line 59, Update the architecture link in the operations documentation to target the existing “Coordination and Self-Healing” heading, or add a matching “Cassandra Lightweight Transactions (LWTs)” heading in the architecture document so the current anchor resolves.docs/v0.6.1/third-party-registries.md-108-114 (1)
108-114: 🔒 Security & Privacy | 🟡 Minor | ⚡ Quick winLimit the permanent-credential claim to NVCF's credential format.
AWS ECR Private supports temporary STS credentials through
aws ecr get-login-password. ECR Public also supports temporary authorization tokens and anonymous pulls. State that NVCF'sACCESS_KEY_ID:SECRET_ACCESS_KEYformat cannot carry a session token, instead of attributing this limitation to ECR.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/third-party-registries.md` around lines 108 - 114, Update the “Adding AWS ECR Registry Credentials” documentation to scope the permanent-credential requirement to NVCF’s ACCESS_KEY_ID:SECRET_ACCESS_KEY format. Clarify that AWS ECR Private supports temporary STS credentials via aws ecr get-login-password, while ECR Public supports temporary authorization tokens and anonymous pulls; avoid implying ECR itself rejects temporary credentials.docs/v0.6.1/function-creation.md-37-38 (1)
37-38: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winCorrect the spelling of
specified.Line 38 says
specificed container image. Replace it withspecified container image.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/function-creation.md` around lines 37 - 38, Correct the typo in the deployment documentation by changing “specificed container image” to “specified container image.”Source: Linters/SAST tools
docs/v0.6.1/container-functions.md-125-127 (1)
125-127: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winFix the missing section link.
[Create the Function]has no destination or reference definition. Link it to the local heading with./container-functions.md#create-the-function.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/container-functions.md` around lines 125 - 127, Update the “Build the Container & Create the Function” paragraph so the “Create the Function” Markdown link targets the local heading using ./container-functions.md#create-the-function.docs/v0.6.1/helm-functions.md-12-16 (1)
12-16: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winRemove Markdown bold from the documentation.
The page uses bold emphasis in the prerequisites, service-name guidance, and API-example heading. Replace the bold spans with plain text or headings.
As per coding guidelines, documentation must use no Markdown bold for emphasis.
Also applies to: 36-40, 42-46
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/helm-functions.md` around lines 12 - 16, Remove all Markdown bold markers from the referenced documentation sections, including the prerequisites, service-name guidance, warning text, and API-example heading; preserve the existing wording and structure while using plain text or appropriate headings without bold emphasis.Source: Coding guidelines
docs/v0.6.1/container-functions.md-14-17 (1)
14-17: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winRemove Markdown bold from the documentation.
The page uses bold emphasis at Lines 14, 17, and 210. Replace those spans with plain text or headings.
As per coding guidelines, documentation must use no Markdown bold for emphasis.
Also applies to: 207-212
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/container-functions.md` around lines 14 - 17, Remove all Markdown bold markers from the documentation text at the referenced passages, including the logging statement and the container root-user warning. Preserve the wording and existing structure while rendering those spans as plain text or suitable headings, and ensure the additional occurrence near the later referenced section is updated as well.Source: Coding guidelines
docs/v0.6.1/control-plane-operations.md-9-18 (1)
9-18: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winApply the repository Markdown formatting rules.
These changed ranges use prohibited Markdown bold, an em dash, or emojis. Normalize
OpenBAOtoOpenBaowhere applicable.
docs/v0.6.1/control-plane-operations.md#L9-L18: remove Markdown bold and the em dash.docs/v0.6.1/runbooks/control-plane-key-rotation-mek.md#L3-L7: remove Markdown bold.docs/v0.6.1/runbooks/overview.md#L4-L8: replace the em dash with ASCII punctuation.docs/v0.6.1/troubleshooting.md#L18-L32: remove Markdown bold and emojis.As per coding guidelines: Documentation must use no Markdown bold for emphasis, no emojis, no em dash, succinct direct sentences, and only standard ASCII in committed text.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/control-plane-operations.md` around lines 9 - 18, Apply the documentation formatting rules across all affected sites: in docs/v0.6.1/control-plane-operations.md lines 9-18, remove bold markup and the em dash and normalize OpenBAO to OpenBao; in docs/v0.6.1/runbooks/control-plane-key-rotation-mek.md lines 3-7, remove bold markup; in docs/v0.6.1/runbooks/overview.md lines 4-8, replace the em dash with ASCII punctuation; and in docs/v0.6.1/troubleshooting.md lines 18-32, remove bold markup and emojis. Keep all committed text ASCII-only, succinct, and direct.Source: Coding guidelines
docs/v0.6.1/troubleshooting.md-814-815 (1)
814-815: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winAdd a language to the shell code fence.
The embedded script fence is missing a language identifier. Use
bashso Markdown tooling and rendered documentation identify the syntax correctly.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/troubleshooting.md` around lines 814 - 815, Update the shell code fence in the troubleshooting documentation by adding the bash language identifier after the opening fence, preserving the embedded script content unchanged.Source: Linters/SAST tools
docs/v0.6.1/troubleshooting.md-627-640 (1)
627-640: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winMake the ConfigMap diagnosis internally consistent.
The expected output lists
cassandra-init-cqlandcassandra-init-script, but the next instruction says the missing ConfigMap iscassandra-migrations. Use the actual resource name consistently so operators know which object to inspect or recreate.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/troubleshooting.md` around lines 627 - 640, Update the ConfigMap troubleshooting section so the expected output, missing-resource guidance, and any inspect/recreate instructions consistently use the actual ConfigMap name; remove the contradictory reference to cassandra-migrations and preserve the documented race-condition diagnosis.docs/v0.6.1/troubleshooting.md-52-52 (1)
52-52: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winReplace the invalid internal references.
The
clean-install-procedurelink points to the current page without a section anchor. TheIncorrect Base64 Docker Credentials Formatreference has no destination. TheRedeploy Stuck Dependenciesreference uses reStructuredText syntax instead of Markdown. Link each reference to an existing heading or document section.Also applies to: 175-179, 480-484
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/troubleshooting.md` at line 52, Replace the invalid internal references in docs/v0.6.1/troubleshooting.md, including the clean-install-procedure, Incorrect Base64 Docker Credentials Format, and Redeploy Stuck Dependencies references at all affected locations. Point each link to its existing heading or document section, use a section anchor for the current-page link, and convert the reStructuredText reference to standard Markdown syntax.docs/v0.6.1/troubleshooting.md-1053-1053 (1)
1053-1053: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winLink to the 0.6.1 cleanup script.
Use the matching
docs/v0.6.1/samples/scripts/force-cleanup-nvcf.shURL. The scripts are currently identical, but the 0.6.1 page should reference its versioned asset.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/troubleshooting.md` at line 1053, Update the force-cleanup-nvcf.sh link in the troubleshooting documentation to reference the version-matched docs/v0.6.1/samples/scripts/force-cleanup-nvcf.sh URL instead of the 0.6.0 asset.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Enterprise
Run ID: 72888127-d50d-41a2-8185-12da02e43651
⛔ Files ignored due to path filters (17)
docs/v0.6.1/images/apipersonalkey.pngis excluded by!**/*.png,!**/*.pngdocs/v0.6.1/images/grpc-reconnect-flow.pngis excluded by!**/*.png,!**/*.pngdocs/v0.6.1/images/grpc-single-client.pngis excluded by!**/*.png,!**/*.pngdocs/v0.6.1/images/lls-self-hosted-arch-streaming-proxy.pngis excluded by!**/*.png,!**/*.pngdocs/v0.6.1/images/low-latency-streaming.pngis excluded by!**/*.png,!**/*.pngdocs/v0.6.1/images/nvcf-banner.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-gpu-cluster-targeting.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-grpc-invocation-path.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-grpc-multicluster-invocation.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-high-level-stack.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-http-invocation-path.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-http-multicluster-invocation.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-llm-invocation-path.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-llm-multicluster-invocation.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-lls-session.pngis excluded by!**/*.png,!**/*.pngdocs/v0.6.1/images/nvcf-multi-region-multi-cluster.svgis excluded by!**/*.svgdocs/v0.6.1/images/self-hosted-min-topology.pngis excluded by!**/*.png,!**/*.png
📒 Files selected for processing (89)
docs/v0.6.1/api.mddocs/v0.6.1/autoscaling/architecture.mddocs/v0.6.1/autoscaling/index.mddocs/v0.6.1/autoscaling/observability.mddocs/v0.6.1/autoscaling/operations.mddocs/v0.6.1/caches.mddocs/v0.6.1/cli.mddocs/v0.6.1/cluster-management/configuration.mddocs/v0.6.1/cluster-management/container-cache.mddocs/v0.6.1/cluster-management/gxcache.mddocs/v0.6.1/cluster-management/index.mddocs/v0.6.1/cluster-management/kai-scheduler.mddocs/v0.6.1/cluster-management/monitoring.mddocs/v0.6.1/cluster-management/reference.mddocs/v0.6.1/cluster-management/self-managed.mddocs/v0.6.1/configure-autoscaling.mddocs/v0.6.1/container-functions.mddocs/v0.6.1/container-tasks.mddocs/v0.6.1/control-plane-installation.mddocs/v0.6.1/control-plane-operations.mddocs/v0.6.1/csp-end-to-end-example-installation.mddocs/v0.6.1/example-dashboards.mddocs/v0.6.1/fake-gpu-operator.mddocs/v0.6.1/function-creation.mddocs/v0.6.1/gateway-routing.mddocs/v0.6.1/generic-http-function-invocation.mddocs/v0.6.1/grpc-function-invocation.mddocs/v0.6.1/grpc-invocation-enablement.mddocs/v0.6.1/grpc-load-test-sli-guide.mddocs/v0.6.1/grpc-load-testing.mddocs/v0.6.1/helm-functions.mddocs/v0.6.1/helm-tasks.mddocs/v0.6.1/helmfile-installation.mddocs/v0.6.1/http-load-test-sli-guide.mddocs/v0.6.1/http-load-testing.mddocs/v0.6.1/http-soak-testing.mddocs/v0.6.1/image-mirroring.mddocs/v0.6.1/index.mddocs/v0.6.1/infrastructure-sizing.mddocs/v0.6.1/installation.mddocs/v0.6.1/llm-function-enablement.mddocs/v0.6.1/llm-gateway.mddocs/v0.6.1/lls-installation.mddocs/v0.6.1/local-development.mddocs/v0.6.1/local-development/multi-cluster-cli.mddocs/v0.6.1/local-development/multi-cluster-helmfile.mddocs/v0.6.1/local-development/single-cluster-cli.mddocs/v0.6.1/local-development/single-cluster-helmfile.mddocs/v0.6.1/manifest.mddocs/v0.6.1/metrics/cassandra/metrics.mddocs/v0.6.1/metrics/ess/metrics.mddocs/v0.6.1/metrics/grpc-proxy/metrics.mddocs/v0.6.1/metrics/init-container/metrics.mddocs/v0.6.1/metrics/invocation-service/metrics.mddocs/v0.6.1/metrics/llm-api-gateway/metrics.mddocs/v0.6.1/metrics/llm-function-invocation-path.mddocs/v0.6.1/metrics/llm-request-router/metrics.mddocs/v0.6.1/metrics/metrics-index.mddocs/v0.6.1/metrics/nvcf-api/metrics.mddocs/v0.6.1/metrics/sis-spot/metrics.mddocs/v0.6.1/metrics/state-metrics/metrics.mddocs/v0.6.1/metrics/utils-container/metrics.mddocs/v0.6.1/metrics/vault-openbao/metrics.mddocs/v0.6.1/observability.mddocs/v0.6.1/optional-enhancements-gpu.mddocs/v0.6.1/optional-enhancements-standalone.mddocs/v0.6.1/optional-enhancements.mddocs/v0.6.1/quickstart.mddocs/v0.6.1/registry-allowlist.mddocs/v0.6.1/release-notes/0.5.0-to-0.6.0-upgrade.mddocs/v0.6.1/release-notes/0.6.0-to-0.6.1-upgrade.mddocs/v0.6.1/release-notes/0.6.0.mddocs/v0.6.1/release-notes/0.6.1.mddocs/v0.6.1/release-notes/index.mddocs/v0.6.1/runbooks/caches.mddocs/v0.6.1/runbooks/control-plane-key-rotation-mek.mddocs/v0.6.1/runbooks/index.mddocs/v0.6.1/runbooks/overview.mddocs/v0.6.1/samples/configs/cp-env-eks-example.yamldocs/v0.6.1/samples/configs/cp-example-secrets.yamldocs/v0.6.1/samples/configs/local-dev-env.yamldocs/v0.6.1/samples/scripts/create-nvcr-pull-secrets.shdocs/v0.6.1/samples/scripts/force-cleanup-nvcf.shdocs/v0.6.1/streaming-functions.mddocs/v0.6.1/task-creation.mddocs/v0.6.1/third-party-registries.mddocs/v0.6.1/troubleshooting.mdfern/docs.ymlfern/versions/v0.6.1.yml
There was a problem hiding this comment.
Caution
Inline review comments failed to post. This is likely due to GitHub's internal server error or limits when posting large numbers of comments. If you are seeing this consistently it is likely a permissions issue. Please check "Moderation" -> "Code review limits" under your organization settings.
Actionable comments posted: 1
Note
Due to the large number of review comments, Critical severity comments were prioritized as inline comments.
🟡 Minor comments (44)
docs/v0.6.1/fake-gpu-operator.md-65-67 (1)
65-67: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winSpecify a language for the error example.
Use
texton the opening fence. The current untyped fence fails MD040.Run the documentation check after this correction. As per coding guidelines, "Run
./tools/ci/check-docsbefore completing documentation changes."🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/fake-gpu-operator.md` around lines 65 - 67, Update the error example code fence in fake-gpu-operator documentation to declare the text language, resolving the MD040 documentation check; then run ./tools/ci/check-docs to verify the documentation change.Sources: Coding guidelines, Linters/SAST tools
docs/v0.6.1/cluster-management/configuration.md-302-369 (1)
302-369: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winFix the nested fenced code blocks.
The unindented closing fences cause the docs checker to treat Lines 352 and 366 as new fenced blocks without a language. Indent each closing fence to match its code block inside the ordered list.
Run the documentation check after this correction. As per coding guidelines, "Run
./tools/ci/check-docsbefore completing documentation changes."🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/cluster-management/configuration.md` around lines 302 - 369, Fix the nested fenced code blocks in the ordered-list steps by indenting each closing fence to match its corresponding opening fence and list indentation, including the YAML, kubectl apply, and kubectl logs blocks. Preserve the existing commands and content, then run ./tools/ci/check-docs to verify the documentation.Sources: Coding guidelines, Linters/SAST tools
docs/v0.6.1/cluster-management/kai-scheduler.md-27-73 (1)
27-73: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winClose the YAML code block correctly.
The docs checker treats the fence at Line 72 as a new code block without a language. Use the MDX fence structure that correctly closes the YAML block inside
<Accordion>.Run the documentation check after this correction. As per coding guidelines, "Run
./tools/ci/check-docsbefore completing documentation changes."🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/cluster-management/kai-scheduler.md` around lines 27 - 73, Correct the Markdown/MDX fence structure in the Accordion containing “kai-scheduler-queues.yaml” so the YAML code block is properly closed rather than starting an untyped block. Preserve the YAML content and run ./tools/ci/check-docs to verify the documentation.Sources: Coding guidelines, Linters/SAST tools
docs/v0.6.1/cluster-management/configuration.md-18-18 (1)
18-18: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winCorrect the node selector description.
Replace
specicwithspecific.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/cluster-management/configuration.md` at line 18, In the node selector description table entry, correct the typo “specic” to “specific” while leaving the rest of the documentation unchanged.Source: Linters/SAST tools
docs/v0.6.1/fake-gpu-operator.md-141-141 (1)
141-141: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winRemove the obsolete
user/path segment.
docs/v0.6.1/fake-gpu-operator.md#L141-L141: change the control-plane installation link to./helmfile-installation.md.docs/v0.6.1/fake-gpu-operator.md#L163-L164: change the self-managed cluster link to./cluster-management/self-managed.md.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/fake-gpu-operator.md` at line 141, Update both links in docs/v0.6.1/fake-gpu-operator.md: change the control-plane installation link at lines 141-141 to ./helmfile-installation.md, and change the self-managed cluster link at lines 163-164 to ./cluster-management/self-managed.md.docs/v0.6.1/cluster-management/configuration.md-32-35 (1)
32-35: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winRemove prohibited Markdown emphasis and non-ASCII punctuation.
docs/v0.6.1/cluster-management/configuration.md#L32-L35: replace Markdown bold with plain text or headings.docs/v0.6.1/cluster-management/container-cache.md#L13-L15: replace Markdown bold with plain text or headings.docs/v0.6.1/cluster-management/gxcache.md#L13-L15: replace Markdown bold with plain text or headings.docs/v0.6.1/cluster-management/kai-scheduler.md#L7-L10: replace Markdown bold with plain text or headings.docs/v0.6.1/cluster-management/reference.md#L17-L25: replace Markdown bold and the em dash with ASCII text.docs/v0.6.1/fake-gpu-operator.md#L7-L10: replace Markdown bold with plain text or headings.docs/v0.6.1/cluster-management/monitoring.md#L10-L12: replace Markdown bold with plain text or headings.As per coding guidelines, "Documentation must use no Markdown bold for emphasis, no emojis, no em dash, succinct direct sentences, and only standard ASCII in committed text."
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/cluster-management/configuration.md` around lines 32 - 35, Remove Markdown bold emphasis from docs/v0.6.1/cluster-management/configuration.md lines 32-35, docs/v0.6.1/cluster-management/container-cache.md lines 13-15, docs/v0.6.1/cluster-management/gxcache.md lines 13-15, docs/v0.6.1/cluster-management/kai-scheduler.md lines 7-10, docs/v0.6.1/fake-gpu-operator.md lines 7-10, and docs/v0.6.1/cluster-management/monitoring.md lines 10-12, replacing it with plain text or headings. In docs/v0.6.1/cluster-management/reference.md lines 17-25, remove Markdown bold and replace the em dash with ASCII text; preserve the existing documentation meaning and use only standard ASCII.Source: Coding guidelines
docs/v0.6.1/samples/scripts/create-nvcr-pull-secrets.sh-7-7 (1)
7-7: 🔒 Security & Privacy | 🟡 Minor | ⚡ Quick winRequire an injected
NGC_API_KEY.Do not instruct users to place an NGC API key in the script. Fail before creating secrets when the environment variable is unset.
Proposed fix
-export NGC_API_KEY="<your-ngc-api-key>" +: "${NGC_API_KEY:?Set NGC_API_KEY before running this script}"🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/samples/scripts/create-nvcr-pull-secrets.sh` at line 7, Update the script’s NGC_API_KEY handling to require the value from the existing environment rather than assigning a placeholder or embedded key. Add an early validation that exits before any secret creation when NGC_API_KEY is unset, while preserving the current secret-generation flow when it is provided.Source: Linters/SAST tools
docs/v0.6.1/local-development.md-8-10 (1)
8-10: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winRemove Markdown bold from the documentation.
The documentation guidelines prohibit Markdown bold in
*.mdfiles. Use plain text or code formatting forlocal development only,Single-cluster,Multi-cluster,CLI, andHelmfile.Also applies to: 33-41, 56-68
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/local-development.md` around lines 8 - 10, Remove Markdown bold formatting from the specified documentation terms in local-development.md, including “local development only,” “Single-cluster,” “Multi-cluster,” “CLI,” and “Helmfile,” while preserving their plain text content and surrounding documentation.Source: Coding guidelines
docs/v0.6.1/quickstart.md-87-101 (1)
87-101: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winList Bazel as a prerequisite for the repository build path.
The documented
bazel buildcommand cannot run unless Bazel or Bazelisk is installed. Add it to the prerequisites, or make the packaged CLI path the default and mark the Bazel path as optional.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/quickstart.md` around lines 87 - 101, Update the prerequisites in the quickstart section for building nvcf-cli so they explicitly require Bazel or Bazelisk before the documented bazel build commands, or clearly label this repository build path as optional and provide the packaged CLI path as the default.docs/v0.6.1/csp-end-to-end-example-installation.md-42-50 (1)
42-50: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winSpecify a language for the numbered code block.
markdownlintreports this fence because it has no language. Usetextfor the installation-order block.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/csp-end-to-end-example-installation.md` around lines 42 - 50, Specify the text language on the numbered installation-order fenced block in the end-to-end installation documentation, changing only the fence annotation while preserving the block contents.Source: Linters/SAST tools
docs/v0.6.1/helmfile-installation.md-194-194 (1)
194-194: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winUse the
0.6.1sample files.This
0.6.1guide links to rawdocs/v0.6.0examples. The repository provides the corresponding0.6.1samples. Link to the local files so the guide remains aligned with this release.Also applies to: 589-591
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/helmfile-installation.md` at line 194, Update the EKS sample references in the v0.6.1 Helmfile installation guide to use the corresponding local 0.6.1 sample files instead of raw docs/v0.6.0 URLs, including the related references around the additional affected lines.docs/v0.6.1/local-development.md-74-88 (1)
74-88: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winComplete the prerequisites for the documented commands.
The local-development pages build
nvcf-cliwith Go and parse verification responses with Python 3, but the prerequisite lists omit those tools.
docs/v0.6.1/local-development.md#L74-L88: add Go to the common prerequisites.docs/v0.6.1/local-development/multi-cluster-cli.md#L40-L59: add Go.docs/v0.6.1/local-development/multi-cluster-cli.md#L308-L316: add Python 3.docs/v0.6.1/local-development/multi-cluster-helmfile.md#L40-L60: add Go.docs/v0.6.1/local-development/multi-cluster-helmfile.md#L250-L257: add Python 3.docs/v0.6.1/local-development/single-cluster-cli.md#L20-L39: add Go.docs/v0.6.1/local-development/single-cluster-cli.md#L266-L273: add Python 3.docs/v0.6.1/local-development/single-cluster-helmfile.md#L20-L40: add Go.docs/v0.6.1/local-development/single-cluster-helmfile.md#L217-L224: add Python 3.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/local-development.md` around lines 74 - 88, Complete the prerequisite lists for the documented commands: add Go to docs/v0.6.1/local-development.md (74-88), docs/v0.6.1/local-development/multi-cluster-cli.md (40-59), docs/v0.6.1/local-development/multi-cluster-helmfile.md (40-60), docs/v0.6.1/local-development/single-cluster-cli.md (20-39), and docs/v0.6.1/local-development/single-cluster-helmfile.md (20-40); add Python 3 to docs/v0.6.1/local-development/multi-cluster-cli.md (308-316), docs/v0.6.1/local-development/multi-cluster-helmfile.md (250-257), docs/v0.6.1/local-development/single-cluster-cli.md (266-273), and docs/v0.6.1/local-development/single-cluster-helmfile.md (217-224).docs/v0.6.1/example-dashboards.md-164-173 (1)
164-173: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winHyphenate the compound phrase.
Change
chicken and egg problemtochicken-and-egg problem.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/example-dashboards.md` around lines 164 - 173, Update the compound phrase in the Step 2 documentation text to use “chicken-and-egg problem” instead of “chicken and egg problem.”Source: Linters/SAST tools
docs/v0.6.1/observability.md-196-196 (1)
196-196: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winHyphenate the compound modifier.
Change
5 minute windowto5-minute window.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/observability.md` at line 196, Update the latency description to hyphenate the compound modifier, changing “5 minute window” to “5-minute window” while preserving the rest of the text.Source: Linters/SAST tools
docs/v0.6.1/metrics/state-metrics/metrics.md-10-12 (1)
10-12: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winRemove the duplicate
nvcf_function_instances_currententry.The same metric and label set appears twice with different descriptions. Keep one canonical definition, or use different metric names if the semantics differ.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/state-metrics/metrics.md` around lines 10 - 12, Remove the duplicate nvcf_function_instances_current row from the metrics table, retaining one canonical definition with its intended description and labels; only introduce a different metric name if the two entries represent distinct semantics.docs/v0.6.1/observability.md-27-32 (1)
27-32: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winUse plain labels or real links in the capability list.
[Metrics Collection]:,[Logging]:,[Tracing]:, and[Dashboards]:have no destinations. In list items, they render as bracketed text rather than links. Remove the brackets or add actual destinations.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/observability.md` around lines 27 - 32, Update the capability list in the observability documentation to remove the unused link-style brackets from Metrics Collection, Logging, Tracing, and Dashboards, unless valid destinations are available; preserve the existing descriptions and list structure.docs/v0.6.1/grpc-load-test-sli-guide.md-50-65 (1)
50-65: 🗄️ Data Integrity & Integration | 🟡 Minor | ⚡ Quick winUse the documented session-init histogram name.
Change the heading to
nvcf_grpc_proxy_service_session_init_secondsand querynvcf_grpc_proxy_service_session_init_seconds_bucket. The current_total_bucketseries is not documented.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/grpc-load-test-sli-guide.md` around lines 50 - 65, Update the session-init metric references in the section for nvcf_grpc_proxy_service_session_init_seconds_total (p95): change the heading to nvcf_grpc_proxy_service_session_init_seconds and use nvcf_grpc_proxy_service_session_init_seconds_bucket in the histogram_quantile query, preserving the existing labels and rate window.docs/v0.6.1/grpc-load-testing.md-7-10 (1)
7-10: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winUse the versioned CLI document for the versioned guide. Change both
../user/cli.mdlinks to./cli.md. Fern maps the v0.6.1 CLI page todocs/v0.6.1/cli.md.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/grpc-load-testing.md` around lines 7 - 10, Update both links in the introductory nvcf-cli setup instructions to use the version-local target ./cli.md instead of ../user/cli.md, preserving their existing link text and surrounding guidance.docs/v0.6.1/observability.md-394-396 (1)
394-396: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winReplace the unrelated redaction reference.
The link targets .NET log redaction documentation, not OpenTelemetry Collector configuration. Link to the Collector Transform or Redaction processor documentation.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/observability.md` around lines 394 - 396, Update the observability documentation’s log redaction reference to point to OpenTelemetry Collector Transform or Redaction processor documentation instead of the unrelated .NET implementation link. Keep the surrounding guidance about collector responsibility and sensitive data redaction unchanged.docs/v0.6.1/metrics/utils-container/metrics.md-8-8 (1)
8-8: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winComplete the
nvcf_worker_service_response_totalrow.Set the type to
Counter, adderror_codeas an interesting label, and describe the total number of service responses processed, including error codes.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/utils-container/metrics.md` at line 8, Update the nvcf_worker_service_response_total row in the metrics documentation: change its type to Counter, add error_code as the interesting label, and describe it as the total number of processed service responses including error codes.docs/v0.6.1/http-soak-testing.md-341-341 (1)
341-341: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winReplace the non-ASCII ellipsis.
The
FUNCTIONSexample contains…. Use an ASCII placeholder such as<id>.As per coding guidelines, documentation must use only standard ASCII in committed text.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/http-soak-testing.md` at line 341, Replace the non-ASCII ellipsis in the FUNCTIONS JSON example with an ASCII placeholder such as <id>, preserving the documented array and funcId structure.Source: Coding guidelines
docs/v0.6.1/http-soak-testing.md-26-31 (1)
26-31: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winRemove Markdown bold from the documentation.
**load_tester_supreme**and**Default control plane sizing:**violate the documentation formatting rule. Use plain text or inline code instead.As per coding guidelines, documentation must use no Markdown bold for emphasis, no emojis, no em dash, succinct direct sentences, and only standard ASCII in committed text.
Also applies to: 313-320
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/http-soak-testing.md` around lines 26 - 31, Remove Markdown bold formatting from the documentation, including load_tester_supreme and Default control plane sizing, while preserving their text and meaning. Use plain text or inline code where code-style emphasis is needed, and ensure the affected sections contain no bold markers or other non-ASCII emphasis characters.Source: Coding guidelines
docs/v0.6.1/metrics/grpc-proxy/metrics.md-27-28 (1)
27-28: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winRemove non-ASCII and bold markup.
**gRPC**violates the no-bold rule.→is not standard ASCII. UsegRPCandgRPC proxy -> NATS pipe.As per coding guidelines, documentation must use no Markdown bold for emphasis, no emojis, no em dash, succinct direct sentences, and only standard ASCII in committed text.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/grpc-proxy/metrics.md` around lines 27 - 28, Update the affected metrics documentation text to remove bold Markdown around gRPC and replace the non-ASCII arrow in “gRPC proxy → NATS pipe” with the ASCII sequence “->”. Preserve the existing wording and meaning while ensuring the committed text uses only standard ASCII.Source: Coding guidelines
docs/v0.6.1/metrics/metrics-index.md-3-3 (1)
3-3: 🗄️ Data Integrity & Integration | 🟡 Minor | ⚡ Quick winAlign the overview claim with the linked pages.
The overview says every linked page lists metric names, types, sources, descriptions, labels, and filters.
docs/v0.6.1/metrics/llm-function-invocation-path.mdlists only metrics and labels. The Vault/OpenBao entry also points to upstream documentation instead of listing the metadata. Change the claim or add the missing metadata.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/metrics-index.md` at line 3, Update the overview sentence in the metrics index so it accurately describes the linked pages, removing metadata claims that are not consistently provided by pages such as llm-function-invocation-path and the Vault/OpenBao entry.docs/v0.6.1/metrics/invocation-service/metrics.md-12-12 (1)
12-12: 🗄️ Data Integrity & Integration | 🟡 Minor | ⚡ Quick winCorrect the
nats_jetstream_publishdescription.The metric name identifies publish operations, but the description says “NATS Stream creation count”. These are different events. Describe the event that the metric actually counts.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/invocation-service/metrics.md` at line 12, Update the description column for the nats_jetstream_publish metric in the metrics table to describe the count of NATS JetStream publish operations, replacing the incorrect stream-creation wording while leaving the other fields unchanged.docs/v0.6.1/metrics/grpc-proxy/metrics.md-32-37 (1)
32-37: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winMake the reproduction command cover the documented metric families.
The command filters only
nvcf_grpc_proxy_*. It does not display the documentedhttp_server_*orrpc_client_*metrics. Remove the filter or include all documented prefixes.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/grpc-proxy/metrics.md` around lines 32 - 37, Update the reproduction curl command in the “Reproducing locally” section so its metrics output includes the documented nvcf_grpc_proxy, http_server, and rpc_client metric families; remove the restrictive grep or expand its prefixes accordingly.docs/v0.6.1/metrics/metrics-index.md-10-10 (1)
10-10: 🗄️ Data Integrity & Integration | 🟡 Minor | ⚡ Quick winAdd gRPC Proxy to Fern navigation. The page exists, but
fern/versions/v0.6.1.ymlhas no corresponding entry.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/metrics-index.md` at line 10, Add the existing gRPC Proxy metrics page to the Fern navigation configuration in the v0.6.1 version manifest, using the established navigation entry format and linking to the metrics page without changing the page content.docs/v0.6.1/http-soak-testing.md-234-237 (1)
234-237: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winAdd the
invoke_latency_msthreshold.The script records
invoke_latency_ms, and the results table requires its p95 to stay below 5,000 ms. Addinvoke_latency_ms: ['p(95)<5000']tooptions.thresholds; otherwise, mark this metric as informational.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/http-soak-testing.md` around lines 234 - 237, Update the options.thresholds configuration to include invoke_latency_ms with a p95 limit below 5,000 ms, matching the results table requirement. If this metric is intentionally informational instead, remove it from the required threshold expectations and document that behavior consistently.docs/v0.6.1/metrics/invocation-service/metrics.md-5-8 (1)
5-8: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winCorrect the metrics table entries.
- Fix the malformed
namespacematcher tonamespace="astro-tenant-nvcf-invocation-service".- Keep the
exported_endpointmatcher. Its backslashes are Markdown escapes and render as^/v2/nvcf.*|^/health$.- Change the
nats_jetstream_publishdescription toNumber of messages published to NATS streams.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/metrics/invocation-service/metrics.md` around lines 5 - 8, Correct the metrics table entries by changing the malformed namespace matcher on app_invocation_error to namespace="astro-tenant-nvcf-invocation-service", while preserving the existing exported_endpoint matcher and its Markdown escapes. Update the nats_jetstream_publish description to “Number of messages published to NATS streams.”docs/v0.6.1/streaming-functions.md-108-110 (1)
108-110: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winCorrect the WebSocket sentence.
Change
to accepting WebSocket connectiontoto accept WebSocket connections.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/streaming-functions.md` around lines 108 - 110, Correct the WebSocket server description in the streaming functions documentation by changing the wording to “to accept WebSocket connections.”Source: Linters/SAST tools
docs/v0.6.1/lls-installation.md-18-20 (1)
18-20: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winApply the documentation formatting rule consistently.
The affected files use Markdown bold for emphasis.
docs/v0.6.1/lls-installation.mdalso uses an em dash. Replace these with plain ASCII text.
docs/v0.6.1/lls-installation.md#L18-L20: remove Markdown bold emphasis.docs/v0.6.1/lls-installation.md#L401-L404: remove Markdown bold and replace the em dash with ASCII punctuation.docs/v0.6.1/streaming-functions.md#L71-L75: remove Markdown bold emphasis.docs/v0.6.1/third-party-registries.md#L7-L10: remove Markdown bold emphasis.As per coding guidelines, documentation must use no Markdown bold for emphasis, no emojis, no em dash, succinct direct sentences, and only standard ASCII in committed text.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/lls-installation.md` around lines 18 - 20, Remove Markdown bold emphasis from docs/v0.6.1/lls-installation.md lines 18-20 and 401-404, replacing the em dash at lines 401-404 with standard ASCII punctuation; also remove Markdown bold emphasis from docs/v0.6.1/streaming-functions.md lines 71-75 and docs/v0.6.1/third-party-registries.md lines 7-10. Preserve the wording while ensuring all affected documentation uses plain ASCII text without bold or em dashes.Source: Coding guidelines
docs/v0.6.1/lls-installation.md-67-68 (1)
67-68: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winCorrect the spelling errors.
Use
Streaminginstead ofStramingandInternalinstead ofInteral.Also applies to: 647-650
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/lls-installation.md` around lines 67 - 68, Correct the spelling errors in the documentation legend: update “Straming” to “Streaming” and “Interal” to “Internal” in the affected descriptions, including the additional occurrence referenced by the comment.Source: Linters/SAST tools
docs/v0.6.1/lls-installation.md-397-402 (1)
397-402: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winCorrect the step references.
The text says to save
TG_ARNfor Step 3, but the value is used in Step 4.2. The next heading jumps from Step 4 to Step 6. Correct the reference and renumber or add Step 5 so the procedure has one clear sequence.Also applies to: 511-513
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/lls-installation.md` around lines 397 - 402, Update the installation procedure’s step references around the TG_ARN instructions: identify the correct step where TG_ARN is consumed (Step 4.2), change the save note to reference that step, and fix the later heading sequence so it progresses continuously from Step 4 through Step 6 by renumbering or adding the missing Step 5.docs/v0.6.1/autoscaling/operations.md-59-59 (1)
59-59: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winFix the broken architecture anchor.
./architecture.md#cassandra-lightweight-transactions-lwtsdoes not match any heading indocs/v0.6.1/autoscaling/architecture.md. Link to the existingCoordination and Self-Healingheading or add the referenced heading.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/autoscaling/operations.md` at line 59, Update the architecture link in the operations documentation to target the existing “Coordination and Self-Healing” heading, or add a matching “Cassandra Lightweight Transactions (LWTs)” heading in the architecture document so the current anchor resolves.docs/v0.6.1/third-party-registries.md-108-114 (1)
108-114: 🔒 Security & Privacy | 🟡 Minor | ⚡ Quick winLimit the permanent-credential claim to NVCF's credential format.
AWS ECR Private supports temporary STS credentials through
aws ecr get-login-password. ECR Public also supports temporary authorization tokens and anonymous pulls. State that NVCF'sACCESS_KEY_ID:SECRET_ACCESS_KEYformat cannot carry a session token, instead of attributing this limitation to ECR.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/third-party-registries.md` around lines 108 - 114, Update the “Adding AWS ECR Registry Credentials” documentation to scope the permanent-credential requirement to NVCF’s ACCESS_KEY_ID:SECRET_ACCESS_KEY format. Clarify that AWS ECR Private supports temporary STS credentials via aws ecr get-login-password, while ECR Public supports temporary authorization tokens and anonymous pulls; avoid implying ECR itself rejects temporary credentials.docs/v0.6.1/function-creation.md-37-38 (1)
37-38: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winCorrect the spelling of
specified.Line 38 says
specificed container image. Replace it withspecified container image.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/function-creation.md` around lines 37 - 38, Correct the typo in the deployment documentation by changing “specificed container image” to “specified container image.”Source: Linters/SAST tools
docs/v0.6.1/container-functions.md-125-127 (1)
125-127: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winFix the missing section link.
[Create the Function]has no destination or reference definition. Link it to the local heading with./container-functions.md#create-the-function.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/container-functions.md` around lines 125 - 127, Update the “Build the Container & Create the Function” paragraph so the “Create the Function” Markdown link targets the local heading using ./container-functions.md#create-the-function.docs/v0.6.1/helm-functions.md-12-16 (1)
12-16: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winRemove Markdown bold from the documentation.
The page uses bold emphasis in the prerequisites, service-name guidance, and API-example heading. Replace the bold spans with plain text or headings.
As per coding guidelines, documentation must use no Markdown bold for emphasis.
Also applies to: 36-40, 42-46
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/helm-functions.md` around lines 12 - 16, Remove all Markdown bold markers from the referenced documentation sections, including the prerequisites, service-name guidance, warning text, and API-example heading; preserve the existing wording and structure while using plain text or appropriate headings without bold emphasis.Source: Coding guidelines
docs/v0.6.1/container-functions.md-14-17 (1)
14-17: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winRemove Markdown bold from the documentation.
The page uses bold emphasis at Lines 14, 17, and 210. Replace those spans with plain text or headings.
As per coding guidelines, documentation must use no Markdown bold for emphasis.
Also applies to: 207-212
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/container-functions.md` around lines 14 - 17, Remove all Markdown bold markers from the documentation text at the referenced passages, including the logging statement and the container root-user warning. Preserve the wording and existing structure while rendering those spans as plain text or suitable headings, and ensure the additional occurrence near the later referenced section is updated as well.Source: Coding guidelines
docs/v0.6.1/control-plane-operations.md-9-18 (1)
9-18: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winApply the repository Markdown formatting rules.
These changed ranges use prohibited Markdown bold, an em dash, or emojis. Normalize
OpenBAOtoOpenBaowhere applicable.
docs/v0.6.1/control-plane-operations.md#L9-L18: remove Markdown bold and the em dash.docs/v0.6.1/runbooks/control-plane-key-rotation-mek.md#L3-L7: remove Markdown bold.docs/v0.6.1/runbooks/overview.md#L4-L8: replace the em dash with ASCII punctuation.docs/v0.6.1/troubleshooting.md#L18-L32: remove Markdown bold and emojis.As per coding guidelines: Documentation must use no Markdown bold for emphasis, no emojis, no em dash, succinct direct sentences, and only standard ASCII in committed text.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/control-plane-operations.md` around lines 9 - 18, Apply the documentation formatting rules across all affected sites: in docs/v0.6.1/control-plane-operations.md lines 9-18, remove bold markup and the em dash and normalize OpenBAO to OpenBao; in docs/v0.6.1/runbooks/control-plane-key-rotation-mek.md lines 3-7, remove bold markup; in docs/v0.6.1/runbooks/overview.md lines 4-8, replace the em dash with ASCII punctuation; and in docs/v0.6.1/troubleshooting.md lines 18-32, remove bold markup and emojis. Keep all committed text ASCII-only, succinct, and direct.Source: Coding guidelines
docs/v0.6.1/troubleshooting.md-814-815 (1)
814-815: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winAdd a language to the shell code fence.
The embedded script fence is missing a language identifier. Use
bashso Markdown tooling and rendered documentation identify the syntax correctly.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/troubleshooting.md` around lines 814 - 815, Update the shell code fence in the troubleshooting documentation by adding the bash language identifier after the opening fence, preserving the embedded script content unchanged.Source: Linters/SAST tools
docs/v0.6.1/troubleshooting.md-627-640 (1)
627-640: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winMake the ConfigMap diagnosis internally consistent.
The expected output lists
cassandra-init-cqlandcassandra-init-script, but the next instruction says the missing ConfigMap iscassandra-migrations. Use the actual resource name consistently so operators know which object to inspect or recreate.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/troubleshooting.md` around lines 627 - 640, Update the ConfigMap troubleshooting section so the expected output, missing-resource guidance, and any inspect/recreate instructions consistently use the actual ConfigMap name; remove the contradictory reference to cassandra-migrations and preserve the documented race-condition diagnosis.docs/v0.6.1/troubleshooting.md-52-52 (1)
52-52: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winReplace the invalid internal references.
The
clean-install-procedurelink points to the current page without a section anchor. TheIncorrect Base64 Docker Credentials Formatreference has no destination. TheRedeploy Stuck Dependenciesreference uses reStructuredText syntax instead of Markdown. Link each reference to an existing heading or document section.Also applies to: 175-179, 480-484
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/troubleshooting.md` at line 52, Replace the invalid internal references in docs/v0.6.1/troubleshooting.md, including the clean-install-procedure, Incorrect Base64 Docker Credentials Format, and Redeploy Stuck Dependencies references at all affected locations. Point each link to its existing heading or document section, use a section anchor for the current-page link, and convert the reStructuredText reference to standard Markdown syntax.docs/v0.6.1/troubleshooting.md-1053-1053 (1)
1053-1053: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick winLink to the 0.6.1 cleanup script.
Use the matching
docs/v0.6.1/samples/scripts/force-cleanup-nvcf.shURL. The scripts are currently identical, but the 0.6.1 page should reference its versioned asset.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/troubleshooting.md` at line 1053, Update the force-cleanup-nvcf.sh link in the troubleshooting documentation to reference the version-matched docs/v0.6.1/samples/scripts/force-cleanup-nvcf.sh URL instead of the 0.6.0 asset.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Enterprise
Run ID: 72888127-d50d-41a2-8185-12da02e43651
⛔ Files ignored due to path filters (17)
docs/v0.6.1/images/apipersonalkey.pngis excluded by!**/*.png,!**/*.pngdocs/v0.6.1/images/grpc-reconnect-flow.pngis excluded by!**/*.png,!**/*.pngdocs/v0.6.1/images/grpc-single-client.pngis excluded by!**/*.png,!**/*.pngdocs/v0.6.1/images/lls-self-hosted-arch-streaming-proxy.pngis excluded by!**/*.png,!**/*.pngdocs/v0.6.1/images/low-latency-streaming.pngis excluded by!**/*.png,!**/*.pngdocs/v0.6.1/images/nvcf-banner.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-gpu-cluster-targeting.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-grpc-invocation-path.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-grpc-multicluster-invocation.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-high-level-stack.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-http-invocation-path.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-http-multicluster-invocation.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-llm-invocation-path.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-llm-multicluster-invocation.svgis excluded by!**/*.svgdocs/v0.6.1/images/nvcf-lls-session.pngis excluded by!**/*.png,!**/*.pngdocs/v0.6.1/images/nvcf-multi-region-multi-cluster.svgis excluded by!**/*.svgdocs/v0.6.1/images/self-hosted-min-topology.pngis excluded by!**/*.png,!**/*.png
📒 Files selected for processing (89)
docs/v0.6.1/api.mddocs/v0.6.1/autoscaling/architecture.mddocs/v0.6.1/autoscaling/index.mddocs/v0.6.1/autoscaling/observability.mddocs/v0.6.1/autoscaling/operations.mddocs/v0.6.1/caches.mddocs/v0.6.1/cli.mddocs/v0.6.1/cluster-management/configuration.mddocs/v0.6.1/cluster-management/container-cache.mddocs/v0.6.1/cluster-management/gxcache.mddocs/v0.6.1/cluster-management/index.mddocs/v0.6.1/cluster-management/kai-scheduler.mddocs/v0.6.1/cluster-management/monitoring.mddocs/v0.6.1/cluster-management/reference.mddocs/v0.6.1/cluster-management/self-managed.mddocs/v0.6.1/configure-autoscaling.mddocs/v0.6.1/container-functions.mddocs/v0.6.1/container-tasks.mddocs/v0.6.1/control-plane-installation.mddocs/v0.6.1/control-plane-operations.mddocs/v0.6.1/csp-end-to-end-example-installation.mddocs/v0.6.1/example-dashboards.mddocs/v0.6.1/fake-gpu-operator.mddocs/v0.6.1/function-creation.mddocs/v0.6.1/gateway-routing.mddocs/v0.6.1/generic-http-function-invocation.mddocs/v0.6.1/grpc-function-invocation.mddocs/v0.6.1/grpc-invocation-enablement.mddocs/v0.6.1/grpc-load-test-sli-guide.mddocs/v0.6.1/grpc-load-testing.mddocs/v0.6.1/helm-functions.mddocs/v0.6.1/helm-tasks.mddocs/v0.6.1/helmfile-installation.mddocs/v0.6.1/http-load-test-sli-guide.mddocs/v0.6.1/http-load-testing.mddocs/v0.6.1/http-soak-testing.mddocs/v0.6.1/image-mirroring.mddocs/v0.6.1/index.mddocs/v0.6.1/infrastructure-sizing.mddocs/v0.6.1/installation.mddocs/v0.6.1/llm-function-enablement.mddocs/v0.6.1/llm-gateway.mddocs/v0.6.1/lls-installation.mddocs/v0.6.1/local-development.mddocs/v0.6.1/local-development/multi-cluster-cli.mddocs/v0.6.1/local-development/multi-cluster-helmfile.mddocs/v0.6.1/local-development/single-cluster-cli.mddocs/v0.6.1/local-development/single-cluster-helmfile.mddocs/v0.6.1/manifest.mddocs/v0.6.1/metrics/cassandra/metrics.mddocs/v0.6.1/metrics/ess/metrics.mddocs/v0.6.1/metrics/grpc-proxy/metrics.mddocs/v0.6.1/metrics/init-container/metrics.mddocs/v0.6.1/metrics/invocation-service/metrics.mddocs/v0.6.1/metrics/llm-api-gateway/metrics.mddocs/v0.6.1/metrics/llm-function-invocation-path.mddocs/v0.6.1/metrics/llm-request-router/metrics.mddocs/v0.6.1/metrics/metrics-index.mddocs/v0.6.1/metrics/nvcf-api/metrics.mddocs/v0.6.1/metrics/sis-spot/metrics.mddocs/v0.6.1/metrics/state-metrics/metrics.mddocs/v0.6.1/metrics/utils-container/metrics.mddocs/v0.6.1/metrics/vault-openbao/metrics.mddocs/v0.6.1/observability.mddocs/v0.6.1/optional-enhancements-gpu.mddocs/v0.6.1/optional-enhancements-standalone.mddocs/v0.6.1/optional-enhancements.mddocs/v0.6.1/quickstart.mddocs/v0.6.1/registry-allowlist.mddocs/v0.6.1/release-notes/0.5.0-to-0.6.0-upgrade.mddocs/v0.6.1/release-notes/0.6.0-to-0.6.1-upgrade.mddocs/v0.6.1/release-notes/0.6.0.mddocs/v0.6.1/release-notes/0.6.1.mddocs/v0.6.1/release-notes/index.mddocs/v0.6.1/runbooks/caches.mddocs/v0.6.1/runbooks/control-plane-key-rotation-mek.mddocs/v0.6.1/runbooks/index.mddocs/v0.6.1/runbooks/overview.mddocs/v0.6.1/samples/configs/cp-env-eks-example.yamldocs/v0.6.1/samples/configs/cp-example-secrets.yamldocs/v0.6.1/samples/configs/local-dev-env.yamldocs/v0.6.1/samples/scripts/create-nvcr-pull-secrets.shdocs/v0.6.1/samples/scripts/force-cleanup-nvcf.shdocs/v0.6.1/streaming-functions.mddocs/v0.6.1/task-creation.mddocs/v0.6.1/third-party-registries.mddocs/v0.6.1/troubleshooting.mdfern/docs.ymlfern/versions/v0.6.1.yml
🛑 Comments failed to post (1)
docs/v0.6.1/streaming-functions.md (1)
131-147: 🎯 Functional Correctness | 🔴 Critical | ⚡ Quick win
Remove the extra closing brace from the function-create payload.
The JSON opens one top-level object. Line 146 closes a second object, so the documented
curlrequest sends invalid JSON.Proposed fix
"functionType": "STREAMING" - } }'📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.-d '{ "name": "'$STREAMING_FUNCTION_NAME'", "inferenceUrl": "/sign_in", "inferencePort": '$STREAMING_SERVER_PORT', "health": { "protocol": "HTTP", "uri": "/v1/streaming/ready", "port": '$CONTROL_SERVER_PORT', "timeout": "PT10S", "expectedStatusCode": 200 }, "containerImage": "'$STREAMING_CONTAINER_IMAGE'", "apiBodyFormat": "CUSTOM", "description": "'$STREAMING_FUNCTION_NAME'", "functionType": "STREAMING" }'🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/v0.6.1/streaming-functions.md` around lines 131 - 147, Remove the extra closing brace at the end of the documented function-create JSON payload, leaving exactly one closing brace for the top-level object so the curl request remains valid JSON.
Signed-off-by: Stephanie Baum <sbaum@nvidia.com>
2 participants
TL;DR
Promote the existing self-managed 0.6.1 documentation from its release-candidate route and make 0.6.1 the default latest stable documentation.
Additional Details
v0.6.1-rctov0.6.1.Latest (0.6.1)and the explicit0.6.1entry use the stable navigation file.For the Reviewer
Review
fern/docs.ymlfor the default version, stable slug, and redirect changes. The frozen documentation files are path-only renames, and the Fern navigation paths follow the new directory.For QA
./tools/ci/check-docspassed with zero errors and the existing advisory version-sync warning becauseimports.yamlis not present in this public checkout.git diff --checkpassed.Customer Release Notes
Not customer visible. This updates the published documentation route for the existing 0.6.1 release.
Plan Summary
Not applicable.
Usage
Not applicable.
Notes
The former release-candidate landing page redirects to the stable 0.6.1 route.
References
None.
Related Pull Requests
None.
Dependencies
None. No license or NOTICE changes.
Issues
Closes #604
Checklist