Skip to content

Comments

fix(update_docs): migrate to code.claude.com domain and refresh docs#5

Open
chrisvaillancourt wants to merge 7 commits intoobra:mainfrom
chrisvaillancourt:fix/update-docs-domain-migration
Open

fix(update_docs): migrate to code.claude.com domain and refresh docs#5
chrisvaillancourt wants to merge 7 commits intoobra:mainfrom
chrisvaillancourt:fix/update-docs-domain-migration

Conversation

@chrisvaillancourt
Copy link

@chrisvaillancourt chrisvaillancourt commented Dec 12, 2025

Summary

  • Updates update_docs.js script to fetch from new code.claude.com domain (previously docs.claude.com which now returns 401)
  • Refreshes all 41 existing documentation files with latest content
  • Adds 5 new documentation pages: claude-code-on-the-web.md, desktop.md, sandboxing.md, slack.md, microsoft-foundry.md
  • Updates SKILL.md with new file references

Motivation and Context

The original script was failing with HTTP 401 because docs.claude.com/llms.txt now requires authentication. Claude Code documentation has migrated to code.claude.com.

This PR:

  1. Fixes the broken update_docs.js script by updating URLs to code.claude.com
  2. Refreshes all documentation with the latest content
  3. Adds newly published documentation pages

How Has This Been Tested?

  • Ran node scripts/update_docs.js successfully (47 files downloaded)
  • Verified all reference files are present and readable

Types of changes

  • Bug fix (non-breaking change which fixes an issue)
  • Documentation update

Checklist

  • My code follows the repository's style guidelines
  • I have added or updated documentation as needed

Related

Supersedes #4 (that PR only fixes the regex pattern but doesn't address the domain migration or 401 auth issue)

Summary by CodeRabbit

  • New Features

    • Added documentation for Claude Code on the web, desktop app, Chrome browser integration (beta), and Slack integration.
    • Added comprehensive guides for plugin discovery, marketplace management, and sandboxing capabilities.
    • Documented new platform support: Microsoft Foundry, cloud execution, and LSP server integration.
  • Documentation

    • Reorganized and expanded all Claude Code documentation with clearer navigation and updated examples.
    • Added new configuration guidance for security, memory management, and deployment options.
    • Improved setup instructions with multi-method installation options and platform-specific guidance.

✏️ Tip: You can customize this high-level summary in your review settings.

- Update LLMS_TXT_URL from docs.claude.com to code.claude.com
- Update CLAUDE_CODE_PATTERN regex for new URL structure
- Update SKILL.md documentation references to new domain
- Add claude_code_docs_map.md to provide hierarchical outline of all docs
- Update script to fetch docs map alongside individual pages
- Document new file in SKILL.md file listing

The docs map provides LLMs with complete structural overview without
reading all 47+ individual documentation files.
Add new documentation files to Quick Reference and file list:
- claude-code-on-the-web.md (web-based Claude Code)
- desktop.md (desktop application)
- sandboxing.md (security isolation)
- slack.md (Slack integration)
- microsoft-foundry.md (Azure AI Foundry)

Remove deprecated migration-guide.md and claude_code_docs_map.md
references.
Refresh all 41 existing documentation files with latest content
from code.claude.com/docs/. Updates include:
- New features and configuration options
- Updated CLI commands and flags
- Improved examples and workflows
- Bug fixes and clarifications
@coderabbitai
Copy link

coderabbitai bot commented Dec 12, 2025

📝 Walkthrough

Walkthrough

This PR comprehensively updates the Claude Code documentation suite, migrating domain references from docs.claude.com to code.claude.com, adding seven new feature documentation pages, restructuring navigation paths across 50+ existing pages, and enhancing documentation infrastructure with auto-generated maps and upstream contribution guidance.

Changes

Cohort / File(s) Summary
Documentation Infrastructure & Script
skills/working-with-claude-code/SKILL.md, skills/working-with-claude-code/scripts/update_docs.js, docs/upstream-improvements-context.md
Updated primary SKILL.md with new quick reference entries for web, desktop, Chrome, sandboxing, and Slack. Modified update_docs.js to fetch from code.claude.com and include documentation map. Added upstream contribution context document detailing domain migration issues and resolutions.
New Feature Documentation
skills/working-with-claude-code/references/claude-code-on-the-web.md, skills/working-with-claude-code/references/desktop.md, skills/working-with-claude-code/references/chrome.md, skills/working-with-claude-code/references/sandboxing.md, skills/working-with-claude-code/references/slack.md, skills/working-with-claude-code/references/microsoft-foundry.md, skills/working-with-claude-code/references/discover-plugins.md
Added comprehensive documentation for Claude Code web platform, desktop app, Chrome integration (beta), sandboxing/security features, Slack integration, Microsoft Foundry deployment, and plugin discovery & installation workflows.
Navigation & Index
skills/working-with-claude-code/references/claude_code_docs_map.md
Added auto-generated hierarchical documentation map with linked sections for Getting started, Build with Claude Code, Administration, Configuration, Reference, Resources, and Deployment.
URL Path & Link Updates
skills/working-with-claude-code/references/amazon-bedrock.md, skills/working-with-claude-code/references/analytics.md, skills/working-with-claude-code/references/checkpointing.md, skills/working-with-claude-code/references/costs.md, skills/working-with-claude-code/references/devcontainer.md, skills/working-with-claude-code/references/google-vertex-ai.md, skills/working-with-claude-code/references/iam.md, skills/working-with-claude-code/references/jetbrains.md, skills/working-with-claude-code/references/legal-and-compliance.md, skills/working-with-claude-code/references/network-config.md, skills/working-with-claude-code/references/security.md, skills/working-with-claude-code/references/terminal-config.md, skills/working-with-claude-code/references/troubleshooting.md
Updated internal documentation links from /en/docs/claude-code/* to /en/* paths. Added navigation footers referencing llms.txt fetch and horizontal rule separators across files. Minor formatting and wording refinements.
Major Documentation Restructuring
skills/working-with-claude-code/references/cli-reference.md, skills/working-with-claude-code/references/common-workflows.md, skills/working-with-claude-code/references/data-usage.md, skills/working-with-claude-code/references/headless.md, skills/working-with-claude-code/references/hooks-guide.md, skills/working-with-claude-code/references/hooks.md, skills/working-with-claude-code/references/interactive-mode.md, skills/working-with-claude-code/references/memory.md, skills/working-with-claude-code/references/overview.md, skills/working-with-claude-code/references/plugins.md, skills/working-with-claude-code/references/plugin-marketplaces.md, skills/working-with-claude-code/references/plugins-reference.md, skills/working-with-claude-code/references/settings.md, skills/working-with-claude-code/references/skills.md, skills/working-with-claude-code/references/sub-agents.md, skills/working-with-claude-code/references/vs-code.md
Extensive content reorganization including rewritten sections, new examples, expanded tables, renamed sections, updated cross-references, and restructured workflows. Added new configuration options, hook events, LSP support, plugin/agent fields, and permission modes. Replaced several overview-based layouts with step-by-step guides.
Provider & Deployment Documentation
skills/working-with-claude-code/references/llm-gateway.md, skills/working-with-claude-code/references/mcp.md, skills/working-with-claude-code/references/model-config.md, skills/working-with-claude-code/references/third-party-integrations.md
Updated provider configuration documentation with new gateway requirements, dynamic MCP server fetching, model alias updates (Opus 4.5), expanded provider comparison table including Microsoft Foundry, and proxy/gateway configuration guidance.
Installation & Quickstart
skills/working-with-claude-code/references/quickstart.md, skills/working-with-claude-code/references/setup.md
Restructured installation methods with multi-tab UI (Native Install, Homebrew, WinGet) replacing single npm approach. Updated authentication guidance, added explicit platform-specific commands, and included auto-update configuration.
Reference & Output Documentation
skills/working-with-claude-code/references/gitlab-ci-cd.md, skills/working-with-claude-code/references/github-actions.md, skills/working-with-claude-code/references/monitoring-usage.md, skills/working-with-claude-code/references/output-styles.md, skills/working-with-claude-code/references/statusline.md
Updated CI/CD integration docs with new model identifiers and bootstrap methods. Enhanced monitoring documentation with normalized metrics/event naming. Expanded output styles with custom Markdown-based styles. Added context window helpers (get\_input\_tokens, get\_output\_tokens, get\_context\_window\_size) and context window usage section.

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~45 minutes


🐰 With whiskers twitched and keyboard keen,
This doc refresh's quite supreme,
From docs.claude to code.claude's gleam,
New features blooming—web, desktop, dream!
Links updated, paths now clean,
Claude Code shines bright—hop on the scene! 🌟

🚥 Pre-merge checks | ✅ 3
✅ Passed checks (3 passed)
Check name Status Explanation
Description Check ✅ Passed Check skipped - CodeRabbit’s high-level summary is enabled.
Title check ✅ Passed The title clearly identifies the main change: migrating documentation source to code.claude.com and refreshing docs. It accurately reflects the primary objective of this PR.
Docstring Coverage ✅ Passed Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

✏️ Tip: You can configure your own custom pre-merge checks in the settings.


Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands and usage tips.

Copy link

@coderabbitai coderabbitai bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actionable comments posted: 13

Note

Due to the large number of review comments, Critical, Major severity comments were prioritized as inline comments.

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (10)
skills/working-with-claude-code/references/output-styles.md (1)

55-88: Frontmatter section: “metadata about the command” looks like a copy/paste (should be “style”)
Line 80-82 says metadata about the “command”, but this section is about output styles. Suggest replacing “command” with “output style”.

-Output style files support frontmatter, useful for specifying metadata about the
-command:
+Output style files support frontmatter, useful for specifying metadata about the
+output style:
skills/working-with-claude-code/references/common-workflows.md (1)

175-218: Fix incorrect documentation path at line 217

The link path /en/sub-agents is incomplete. The correct path is /en/docs/claude-code/sub-agents. Update the link:

- [subagents documentation](/en/sub-agents)
+ [subagents documentation](/en/docs/claude-code/sub-agents)
skills/working-with-claude-code/references/amazon-bedrock.md (1)

175-201: Confirm IAM permissions are valid, but scope should be narrowed per AWS best practices.

The IAM policy additions are correct and necessary:

  • bedrock:InvokeModel and bedrock:InvokeModelWithResponseStream are required for model invocation
  • bedrock:ListInferenceProfiles supports inference profile discovery
  • aws-marketplace:Subscribe and ViewSubscriptions are required for serverless Marketplace model enablement
  • The bedrock.amazonaws.com condition on Marketplace actions is an appropriate security measure

However, AWS Bedrock documentation recommends scoping these permissions to specific models and regions rather than using wildcards. Consider narrowing Resource to target specific model ARNs (e.g., arn:aws:bedrock:us-east-1:*:foundation-model/anthropic.claude-3-5-sonnet-*) and constraining Marketplace actions to applicable regions where Marketplace models are used.

skills/working-with-claude-code/references/plugin-marketplaces.md (1)

250-287: Address markdownlint MD040 (missing fenced code language)
The linter indicates at least one fenced block is missing a language tag—please update that fence to shell, json, or text as appropriate so CI stays clean.

skills/working-with-claude-code/references/google-vertex-ai.md (1)

128-134: Use docs.anthropic.com instead of docs.claude.com for public-facing documentation links

The link to docs.claude.com may require authentication (Console login or SSO). For publicly accessible Claude documentation, migrate to the canonical URL at docs.anthropic.com. Note: code.claude.com is the Claude Code product interface, not documentation—do not use it as a documentation source.

skills/working-with-claude-code/references/plugins.md (1)

124-137: Add language to the plugin tree code block (MD040).

-```
+```text
 my-first-plugin/
 ├── .claude-plugin/
 │   └── plugin.json          # Plugin metadata
 ...
     └── hooks.json

</blockquote></details>
<details>
<summary>skills/working-with-claude-code/references/model-config.md (1)</summary><blockquote>

`52-59`: **Add language to settings example (and fix indentation consistency).**  


```diff
-```
+```json
 {
-    "permissions": {
-        ...
-    },
-    "model": "opus"
+  "permissions": {
+    ...
+  },
+  "model": "opus"
 }

</blockquote></details>
<details>
<summary>skills/working-with-claude-code/references/third-party-integrations.md (1)</summary><blockquote>

`184-187`: **Doc inconsistency: SKIP_AUTH flags list omits Foundry.**  
You introduce `CLAUDE_CODE_SKIP_FOUNDRY_AUTH` earlier, but the summary sentence only names Bedrock/Vertex.  


```diff
-Claude Code uses the `ANTHROPIC_AUTH_TOKEN` for the `Authorization` header when needed. The `SKIP_AUTH` flags (`CLAUDE_CODE_SKIP_BEDROCK_AUTH`, `CLAUDE_CODE_SKIP_VERTEX_AUTH`) are used in LLM gateway scenarios where the gateway handles provider authentication.
+Claude Code uses the `ANTHROPIC_AUTH_TOKEN` for the `Authorization` header when needed. The `SKIP_AUTH` flags (`CLAUDE_CODE_SKIP_BEDROCK_AUTH`, `CLAUDE_CODE_SKIP_VERTEX_AUTH`, `CLAUDE_CODE_SKIP_FOUNDRY_AUTH`) are used in LLM gateway scenarios where the gateway handles provider authentication.
skills/working-with-claude-code/references/plugins-reference.md (1)

359-377: Broken markdown fence around “Version management” (will corrupt rendering). There’s a ```json opening fence with no JSON content/closure before “## See also”. This is likely to break the rest of the page.

Suggested fix (either add the missing JSON example, or remove the stray fence). For example, if no code block is intended:

-```json  theme={null}
-
-## See also
+## See also
skills/working-with-claude-code/references/settings.md (1)

32-57: Fix code fence language tag (“```JSON Example …” isn’t a valid fence language for most linters).

-```JSON Example settings.json theme={null}
+```json  theme={null}
 {
   "permissions": {
     ...
   }
 }
🟡 Minor comments (52)
skills/working-with-claude-code/references/headless.md-203-209 (1)

203-209: Fix markdownlint MD034: bare URL at Line 209. Wrap the URL in a proper link (or angle brackets) to satisfy no-bare-urls.

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the llms.txt file at: <https://code.claude.com/docs/llms.txt>
skills/working-with-claude-code/references/legal-and-compliance.md-39-41 (1)

39-41: Fix markdownlint violations: use consistent HR and avoid bare URL.

This footer introduces a different horizontal rule style than the rest of the file (Line 34 uses ***) and uses a bare URL.

----
+***
 
-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/output-styles.md-30-37 (1)

30-37: Potential ambiguity/overreach in “exclude instructions” guarantees
Bullets read like hard guarantees (“All output styles exclude…”, “All output styles trigger reminders…”). If there are versions/configs where this differs, consider softening to “typically” / “by default”, or add a clarifying caveat/version note.

skills/working-with-claude-code/references/output-styles.md-114-114 (1)

114-114: Fix markdownlint MD034 (no-bare-urls) for llms.txt link
Use an autolink or a normal Markdown link.

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the llms.txt file at: <https://code.claude.com/docs/llms.txt>
skills/working-with-claude-code/references/devcontainer.md-9-14 (1)

9-14: Tighten the warning sentence to avoid ambiguity about “anything accessible”.
Current wording could be read as “exfiltrate anything (globally)” vs “anything accessible inside the container / through allowed egress”. Consider a small clarification for precision.

-  When executed with `--dangerously-skip-permissions`, devcontainers don't prevent a malicious project from exfiltrating anything accessible in the devcontainer including Claude Code credentials.
+  When executed with `--dangerously-skip-permissions`, devcontainers don't prevent a malicious project from exfiltrating data accessible from within the devcontainer (including Claude Code credentials).
skills/working-with-claude-code/references/security.md-90-90 (1)

90-90: Replace “See here” with descriptive link text (markdownlint MD059).

Line 90 currently uses non-descriptive link text. Suggest:

-See [here](/en/vs-code#security) for more information on the security of running Claude Code in an IDE.
+See [IDE security for Claude Code in VS Code](/en/vs-code#security) for more information.
skills/working-with-claude-code/references/security.md-141-141 (1)

141-141: Avoid bare URL; use a proper Markdown link (markdownlint MD034).

Line 141:

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch [llms.txt](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/sandboxing.md-139-147 (1)

139-147: Replace bare URL with a proper Markdown link

MD034 is flagging the bare URL; wrap it as a link (and optionally make the anchor more descriptive).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the documentation index at: [code.claude.com/docs/llms.txt](https://code.claude.com/docs/llms.txt)

Also applies to: 224-224

skills/working-with-claude-code/references/sandboxing.md-87-95 (1)

87-95: Fix markdownlint “descriptive link text” around the Settings reference

Line 87-90 reads like the link target is “Settings”, but the sentence is already “See Settings…”, which can trip MD059 depending on config. I’d make the link text carry the meaning and avoid generic “See … for …”.

-Customize sandbox behavior through your `settings.json` file. See [Settings](/en/settings#sandbox-settings) for complete configuration reference.
+Customize sandbox behavior through your `settings.json` file. See [sandbox settings reference](/en/settings#sandbox-settings) for the complete configuration.
skills/working-with-claude-code/references/terminal-config.md-70-71 (1)

70-71: Remove consecutive blank lines before the separator and footer

Lines 70-71 contain multiple consecutive blank lines before the --- separator, and line 73 has an extra blank line after it. Collapse these to single blank lines to match markdown style conventions.

skills/working-with-claude-code/references/claude-code-on-the-web.md-130-137 (1)

130-137: Fix fenced block missing language (markdownlint MD040).

Line 133 uses a fenced code block without a language. Consider dotenv (or text) to satisfy MD040 and improve readability.

-  ```
+  ```dotenv
   API_KEY=your_api_key
   DEBUG=true

</blockquote></details>
<details>
<summary>skills/working-with-claude-code/references/claude-code-on-the-web.md-489-489 (1)</summary><blockquote>

`489-489`: **Replace bare URL to satisfy markdownlint MD034.** 

Line 489 is a bare URL; wrap it in a markdown link.

```diff
-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the llms.txt file at: [https://code.claude.com/docs/llms.txt](https://code.claude.com/docs/llms.txt)
skills/working-with-claude-code/references/claude-code-on-the-web.md-141-159 (1)

141-159: Review hook command quoting against actual hook execution semantics.

The example shows "command": "\"$CLAUDE_PROJECT_DIR\"/scripts/install_pkgs.sh" with embedded escaped quotes. The safety concern depends on whether the hook runner invokes this via shell (where quotes get interpreted) or via direct execve (where quotes become literal characters). Without access to the hook execution implementation, the suggested safer alternative is to avoid ambiguous quoting: either show a plain unquoted path if the runner handles variable expansion directly, or explicitly wrap the full command in bash -c if shell interpretation is expected.

skills/working-with-claude-code/references/vs-code.md-144-144 (1)

144-144: Format bare URL as markdown link (MD034 violation).

The URL on line 144 is a bare URL and violates the no-bare-urls rule. Format it as a proper markdown link.

Apply this diff to fix the violation:

-  > To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+  > To find navigation and other pages in this documentation, fetch the llms.txt file at: [https://code.claude.com/docs/llms.txt](https://code.claude.com/docs/llms.txt)
skills/working-with-claude-code/references/troubleshooting.md-148-155 (1)

148-155: Minor wording nit: “Enterprise-managed” hyphenation + Windows home dir example.

LanguageTool is right that “Enterprise-managed” reads better hyphenated. Also, Windows ~ resolution depends on shell; consider “typically C:\Users\YourName” (you already say “such as”, so just hyphenate).

skills/working-with-claude-code/references/troubleshooting.md-363-365 (1)

363-365: Fix MD034: bare URL in footer.

markdownlint will keep flagging this. Use a proper Markdown link or wrap the URL in angle brackets. Example:

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the llms.txt file at: <https://code.claude.com/docs/llms.txt>
skills/working-with-claude-code/references/skills.md-612-612 (1)

612-612: Fix markdownlint MD034 (bare URL).

Line 612 is a bare URL; wrap it in angle brackets (or make it a normal markdown link).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the llms.txt file at: <https://code.claude.com/docs/llms.txt>
skills/working-with-claude-code/references/common-workflows.md-266-266 (1)

266-266: Grammar fix: “create” → “creates”
Line 266: “Claude analyzes … and create …” should be “and creates …”.

-Claude analyzes the current implementation and create a comprehensive plan. Refine with follow-ups:
+Claude analyzes the current implementation and creates a comprehensive plan. Refine with follow-ups:
skills/working-with-claude-code/references/monitoring-usage.md-446-446 (1)

446-446: Fix LanguageTool EN_WORD_COHERENCY (analyze vs analyse)
Pick one variant and use it consistently within the doc.

skills/working-with-claude-code/references/memory.md-90-198 (1)

90-198: Fix markdownlint MD040 (add languages to fenced blocks)
The directory-tree examples should use a language (typically text) to satisfy MD040.

-```
+```text
 your-project/
 ├── .claude/
 │   ├── CLAUDE.md           # Main project instructions
 │   └── rules/
 │       ├── code-style.md   # Code style guidelines
 │       ├── testing.md      # Testing conventions
 │       └── security.md     # Security requirements

</blockquote></details>
<details>
<summary>skills/working-with-claude-code/references/memory.md-200-207 (1)</summary><blockquote>

`200-207`: **Fix markdownlint MD007 (list indentation inside `<Tip>`)**  
Either unindent the list items to column 0, or convert to a plain paragraph list without nested indentation so markdownlint passes.

</blockquote></details>
<details>
<summary>skills/working-with-claude-code/references/data-usage.md-99-101 (1)</summary><blockquote>

`99-101`: **Fix markdownlint MD034 (no-bare-urls) in llms.txt footer**  


```diff
-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/memory.md-226-228 (1)

226-228: Fix markdownlint MD034 (no-bare-urls) in llms.txt footer

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/gitlab-ci-cd.md-465-467 (1)

465-467: Fix markdownlint MD034 (no-bare-urls) in llms.txt footer

 > To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).

Committable suggestion skipped: line range outside the PR's diff.

skills/working-with-claude-code/references/github-actions.md-14-16 (1)

14-16: All claims verified; consider adding maintenance flag for future updates
The model ID claude-opus-4-5-20251101 and "defaults to Sonnet" claim are both current and accurate per official Claude Code GitHub Actions documentation. However, this is inherently time-sensitive content (model availability, defaults, model IDs). Consider adding a comment or label to flag this section for periodic review as Anthropic releases new models or changes defaults.

skills/working-with-claude-code/references/mcp.md-206-206 (1)

206-206: Fix hyphenation: "open source" → "open-source".

When "open source" is used as a compound adjective before a noun, it should be hyphenated.

-Claude Code can connect to hundreds of external tools and data sources through the [Model Context Protocol (MCP)](https://modelcontextprotocol.io/introduction), an open source standard for AI-tool integrations.
+Claude Code can connect to hundreds of external tools and data sources through the [Model Context Protocol (MCP)](https://modelcontextprotocol.io/introduction), an open-source standard for AI-tool integrations.
skills/working-with-claude-code/references/sub-agents.md-155-155 (1)

155-155: Fix incomplete table cell description.

Line 155 starts with "Can be a model alias" but is missing a subject. Rework as: "Model to use; can be a model alias..." or similar.

-| `model`          | No       | Model to use for this subagent. Can be a model alias (`sonnet`, `opus`, `haiku`) or `'inherit'` to use the main conversation's model. If omitted, defaults to the [configured subagent model](/en/model-config) |
+| `model`          | No       | Specifies which model the subagent uses. Can be a model alias (`sonnet`, `opus`, `haiku`) or `'inherit'` to use the main conversation's model. If omitted, defaults to the [configured subagent model](/en/model-config) |
skills/working-with-claude-code/references/mcp.md-6-6 (1)

6-6: Remove unused variable declaration.

Line 6 declares ifYouAreAnAiAgentReadingThisYouShouldInsteadFetch but it's never referenced. Remove this line to clean up the code.

- const ifYouAreAnAiAgentReadingThisYouShouldInsteadFetch = 'https://api.anthropic.com/mcp-registry/docs';
skills/working-with-claude-code/references/desktop.md-41-46 (1)

41-46: Add a language to fenced code block (MD040).
This is a pattern file; gitignore (or text) would be fine.

-```
+```gitignore
 .env
 .env.local
 .env.*
 **/.claude/settings.local.json

</blockquote></details>
<details>
<summary>skills/working-with-claude-code/references/setup.md-26-56 (1)</summary><blockquote>

`26-56`: **Remove duplicated `theme={null}` attributes in fenced blocks (likely generator bug).**  
Many code fences have `theme={null} theme={null}` which looks accidental and can break rendering/tooling.  


```diff
-    ```sh theme={null} theme={null}
+    ```sh theme={null}
     brew install --cask claude-code
     ```
...
-    ```bash theme={null} theme={null}
+    ```bash theme={null}
     curl -fsSL https://claude.ai/install.sh | bash
     ```
...
-    ```powershell theme={null} theme={null}
+    ```powershell theme={null}
     irm https://claude.ai/install.ps1 | iex
     ```
...
-    ```batch theme={null} theme={null}
+    ```batch theme={null}
     curl -fsSL https://claude.ai/install.cmd -o install.cmd && install.cmd && del install.cmd
     ```
...
-    ```sh theme={null} theme={null}
+    ```sh theme={null}
     npm install -g @anthropic-ai/claude-code
     ```
skills/working-with-claude-code/references/iam.md-199-202 (1)

199-202: Avoid bare URL in footer (MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/desktop.md-83-85 (1)

83-85: Avoid bare URL in footer (MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/plugins.md-394-396 (1)

394-396: Avoid bare URL in footer (MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/setup.md-166-172 (1)

166-172: Use a heading for “Binary integrity and code signing” (don’t style bold text as a section title).
This triggers markdownlint MD036 and also makes navigation/anchors inconsistent.

-**Binary integrity and code signing**
+### Binary integrity and code signing
skills/working-with-claude-code/references/model-config.md-132-134 (1)

132-134: Avoid bare URL in footer (MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/microsoft-foundry.md-110-112 (1)

110-112: Avoid bare URL in footer (MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/desktop.md-54-59 (1)

54-59: Fix incomplete sentence: “This is useful for:” has no list.
Currently reads like content was dropped. Either add bullets or remove the lead-in.

-From the desktop app, you can kick off Claude Code sessions that run on Anthropic's secure cloud infrastructure. This is useful for:
-
-To start a web session from desktop, select a remote environment when creating a new session.
+From the desktop app, you can kick off Claude Code sessions that run on Anthropic's secure cloud infrastructure.
+
+To start a web session from desktop, select a remote environment when creating a new session.
skills/working-with-claude-code/references/microsoft-foundry.md-30-44 (1)

30-44: Promote “Option A/B” to headings (instead of bold lead-ins).
Improves navigation and fixes MD036.

-**Option A: API key authentication**
+#### Option A: API key authentication
...
-**Option B: Microsoft Entra ID authentication**
+#### Option B: Microsoft Entra ID authentication
skills/working-with-claude-code/references/claude_code_docs_map.md-1015-1017 (1)

1015-1017: Avoid bare URL in footer (MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/third-party-integrations.md-263-265 (1)

263-265: Avoid bare URL in footer (MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/setup.md-303-305 (1)

303-305: Avoid bare URL in footer (MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/slash-commands.md-354-356 (1)

354-356: Add a language to the fenced block (markdownlint MD040).

-```
+```text
 > Run /write-unit-test when you are about to start writing tests.
skills/working-with-claude-code/references/settings.md-522-524 (1)

522-524: Avoid bare URL in footer (markdownlint MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/llm-gateway.md-17-29 (1)

17-29: Use a real heading instead of bold label (markdownlint MD036).

-**API format**
+### API format
skills/working-with-claude-code/references/quickstart.md-331-333 (1)

331-333: Avoid bare URL in footer (markdownlint MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/cli-reference.md-130-146 (1)

130-146: Avoid bare URL in footer if markdownlint is enforced repo-wide.

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/slash-commands.md-511-513 (1)

511-513: Avoid bare URL in footer (markdownlint MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/overview.md-121-123 (1)

121-123: Avoid bare URL in footer (markdownlint MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/hooks.md-1131-1133 (1)

1131-1133: Avoid bare URL in footer (markdownlint MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/llm-gateway.md-173-175 (1)

173-175: Avoid bare URL in footer (markdownlint MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/plugins-reference.md-380-382 (1)

380-382: Avoid bare URL in footer (markdownlint MD034).

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/slash-commands.md-201-203 (1)

201-203: Update the Models overview URL to the current path. The link at line 201 references an outdated path pattern. Change https://docs.claude.com/en/docs/about-claude/models/overview to https://docs.claude.com/claude/docs/models-overview.

🧹 Nitpick comments (24)
skills/working-with-claude-code/references/devcontainer.md (2)

73-77: Link updates look correct; consider consistency of “Enterprise …” label.
The new /en/security and /en/network-config paths align with the domain/path migration described in the PR. Minor: if this page is not enterprise-specific, “Network configuration” may be a more generally accurate label.


80-82: Avoid hardcoding the full llms.txt URL (prefer relative) to keep forks/envs working.
Hardcoding https://code.claude.com/docs/llms.txt can break in previews, forks, or alternate hosting. Prefer a relative /docs/llms.txt (or ./llms.txt depending on hosting layout) unless you explicitly want to pin to the canonical host.

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the llms.txt file at: /docs/llms.txt
skills/working-with-claude-code/references/security.md (1)

23-27: Optional: make list punctuation consistent (periods / trailing punctuation).

Some bullets end with periods while others don’t (e.g., Line 23 vs Lines 96–103). Consider standardizing for readability.

Also applies to: 96-103

skills/working-with-claude-code/references/sandboxing.md (1)

94-95: Optional editorial cleanups from LanguageTool

These are small style/readability wins (and may reduce lint noise if you gate on LanguageTool):

-  * `docker` is incompatible with running in the sandbox. Consider specifying `docker` in `excludedCommands` to force it to run outside of the sandbox.
+  * `docker` is incompatible with running in the sandbox. Consider specifying `docker` in `excludedCommands` to force it to run outside the sandbox.

-  You can disable this escape hatch by setting `"allowUnsandboxedCommands": false` in your [sandbox settings](/en/settings#sandbox-settings). When disabled, the `dangerouslyDisableSandbox` parameter is completely ignored and all commands must run sandboxed or be explicitly listed in `excludedCommands`.
+  You can disable this escape hatch by setting `"allowUnsandboxedCommands": false` in your [sandbox settings](/en/settings#sandbox-settings). When disabled, the `dangerouslyDisableSandbox` parameter is ignored and all commands must run sandboxed or be explicitly listed in `excludedCommands`.

-* Privilege Escalation via Unix Sockets: The `allowUnixSockets` configuration can inadvertently grant access to powerful system services that could lead to sandbox bypasses.
+* Privilege Escalation via Unix Sockets: The `allowUnixSockets` configuration can grant access to powerful system services that could lead to sandbox bypasses.

-The sandbox runtime is available as an open source npm package for use in your own agent projects.
+The sandbox runtime is available as an open-source npm package for use in your own agent projects.

-* **Compatibility**: Some tools that require specific system access patterns may need configuration adjustments, or may even need to be run outside of the sandbox
+* **Compatibility**: Some tools that require specific system access patterns may need configuration adjustments, or may need to be run outside the sandbox

Also applies to: 100-101, 156-158, 200-213

skills/working-with-claude-code/references/jetbrains.md (1)

151-153: Footer is helpful; consider avoiding hard-coded domain to reduce future churn.
If your docs renderer supports it, prefer a relative link (or at least a markdown link) rather than embedding https://code.claude.com/... in plain text.

skills/working-with-claude-code/references/terminal-config.md (1)

72-74: Footer navigation: make link style consistent + avoid hard-coded URL if you support multiple bases
The footer uses a raw absolute URL while the rest uses markdown links. Consider formatting as a standard markdown link (and optionally reference-style) to keep the docs consistent and easier to update.

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch: [llms.txt](https://code.claude.com/docs/llms.txt)
skills/working-with-claude-code/references/claude-code-on-the-web.md (1)

218-444: Allowlist section mixes “domains” with URLs/escapes; consider normalizing to hostnames.

This section is titled “domains” but includes entries like [www.github.com](http://www.github.com) and escaped strings like raw\.githubusercontent.com. For readers configuring allowlists, hostnames-only (and unescaped) is less ambiguous (e.g., www.github.com, raw.githubusercontent.com). If you need to indicate “HTTP also works”, call it out in prose rather than encoding it as links.

skills/working-with-claude-code/references/troubleshooting.md (2)

52-54: WSL guidance: warning is good, but consider making the “don’t do this” alternatives more actionable.

The <Warning> advises against appendWindowsPath = false / uninstalling Windows Node. Consider adding a concrete “safe next step” (e.g., point readers to preferred PATH/nvm setup section immediately) so people don’t stop at “don’t”.


214-215: Environment variable guidance: consider stating when restart is required.

Setting USE_BUILTIN_RIPGREP=0 usually needs a new shell / restart of Claude Code to take effect; adding that sentence would reduce confusion.

skills/working-with-claude-code/references/slack.md (1)

211-211: Make the llms.txt reference consistently formatted (and preferably clickable) across docs. If other pages use a standardized footer snippet, consider matching it exactly to avoid churn in future automated refreshes.

skills/working-with-claude-code/scripts/update_docs.js (2)

24-38: Add timeout + redirect handling to fetchUrl (external-call reliability)
Right now a hung socket can stall forever, and a 301/302 from future migrations will hard-fail. Consider switching to built-in fetch() (Node 18+) with an AbortController, or add req.setTimeout(...) + basic redirect support.

 function fetchUrl(url) {
   return new Promise((resolve, reject) => {
-    https.get(url, (res) => {
+    const req = https.get(url, (res) => {
       let data = '';
       res.on('data', (chunk) => data += chunk);
       res.on('end', () => {
-        if (res.statusCode >= 200 && res.statusCode < 300) {
+        // Follow redirects (common during doc hosting migrations)
+        if (res.statusCode >= 300 && res.statusCode < 400 && res.headers.location) {
+          return resolve(fetchUrl(res.headers.location));
+        }
+        if (res.statusCode >= 200 && res.statusCode < 300) {
           resolve(data);
         } else {
           reject(new Error(`HTTP ${res.statusCode}: ${res.statusMessage}`));
         }
       });
-    }).on('error', reject);
+    });
+    req.setTimeout(30_000, () => req.destroy(new Error('Request timed out')));
+    req.on('error', reject);
   });
 }

90-102: Docs map fetch: consider failing fast (or at least surfacing in summary)
If the map is intended to be “required,” you probably want to exit non-zero when mapResult.success === false; otherwise users may assume “✅ complete” while the navigation index is missing.

skills/working-with-claude-code/references/google-vertex-ai.md (1)

162-164: Prefer linkified llms.txt footer (consistency + avoids MD034 elsewhere)

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/data-usage.md (1)

56-57: Large inline <img>: consider adding a simple Markdown fallback link
Optional, but improves readability in renderers that strip HTML attributes.

skills/working-with-claude-code/references/github-actions.md (1)

675-677: Prefer linkified llms.txt footer (consistency; avoids MD034 in other files)

-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt).
skills/working-with-claude-code/references/monitoring-usage.md (1)

85-86: OTEL_LOG_USER_PROMPTS: consider a stronger warning callout
Optional, but since this can capture sensitive data, a short warning block near the env var table could prevent accidental leakage.

skills/working-with-claude-code/references/sub-agents.md (1)

278-278: Add language specifications to code fence blocks.

Several code blocks (lines 278, 304, 345, 355, 511, 519) use fenced code blocks without specifying a language. Add language identifiers for syntax highlighting:

-```
+```

Lines affected: 278 (example scenario), 304 (example scenario), 345 (user example), 355 (user example), 511 (initial invocation), 519 (resume agent command).

Also applies to: 304-304, 345-345, 355-355, 511-511, 519-519

skills/working-with-claude-code/references/costs.md (1)

38-38: Consider more concise phrasing for clarity.

"In order to get cost metrics, several large enterprises..." could be simplified to "To get cost metrics..." The phrase "in order to" is redundant in this context.

-On Bedrock and Vertex, Claude Code does not send metrics from your cloud. In order to get cost metrics, several large enterprises reported using [LiteLLM](/en/third-party-integrations#litellm), which is an open-source tool that helps companies [track spend by key](https://docs.litellm.ai/docs/proxy/virtual_keys#tracking-spend). This project is unaffiliated with Anthropic and we have not audited its security.
+On Bedrock and Vertex, Claude Code does not send metrics from your cloud. To get cost metrics, several large enterprises reported using [LiteLLM](/en/third-party-integrations#litellm), which is an open-source tool that helps companies [track spend by key](https://docs.litellm.ai/docs/proxy/virtual_keys#tracking-spend). This project is unaffiliated with Anthropic and we have not audited its security.
skills/working-with-claude-code/references/hooks-guide.md (1)

330-333: Refactor repetitive sentence beginnings in "Learn more" section.

Three consecutive sentences begin with "For", creating awkward repetition. Consider combining or restructuring:

-* For reference documentation on hooks, see [Hooks reference](/en/hooks).
-* For comprehensive security best practices and safety guidelines, see [Security Considerations](/en/hooks#security-considerations) in the hooks reference documentation.
-* For troubleshooting steps and debugging techniques, see [Debugging](/en/hooks#debugging) in the hooks reference documentation.
+* Reference documentation is available in [Hooks reference](/en/hooks).
+* Security best practices and safety guidelines are covered in [Security Considerations](/en/hooks#security-considerations).
+* Troubleshooting and debugging techniques can be found in [Debugging](/en/hooks#debugging).
skills/working-with-claude-code/references/claude_code_docs_map.md (1)

7-9: If this is generated, consider failing the workflow when required fields are missing (e.g., title).
Prevents future “null” regressions from silently landing.

skills/working-with-claude-code/references/model-config.md (1)

21-28: Avoid “currently …” for model alias mappings unless generation guarantees freshness.
These values are time-sensitive; consider either removing “currently …” or clarifying it’s “as of ”.

skills/working-with-claude-code/references/overview.md (1)

66-67: Wording/link update is fine, but “host on AWS or GCP” phrasing is slightly inconsistent with the new “Build with the Agent SDK” card. Consider aligning terminology between the bullet and the cards to avoid mixed messaging.

skills/working-with-claude-code/references/settings.md (1)

190-210: Precedence example is good; consider clarifying “enterprise managed policies can’t be overridden” applies even to CLI flags. (Currently implied, but making it explicit would avoid misreads.)

skills/working-with-claude-code/references/hooks.md (1)

961-964: Python example uses True in JSON output—this is correct for Python dicts, but readers may confuse it with JSON. Consider explicitly calling it “Python dict (serialized to JSON)” or show the emitted JSON equivalent.

Comment on lines +51 to +58
**Option D: AWS Management Console credentials**

```bash theme={null}
aws login
```

[Learn more](https://docs.aws.amazon.com/signin/latest/userguide/command-line-sign-in.html) about `aws login`.

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🔴 Critical

🧩 Analysis chain

🌐 Web query:

aws login command AWS CLI credential method

💡 Result:

Short answer: there is no single built‑in "aws login" command for all credential types. Common AWS CLI credential methods:

  • Static keys via aws configure (stores aws_access_key_id / aws_secret_access_key). [1]
  • IAM Identity Center (SSO): run aws configure sso to create a profile, then aws sso login --profile to fetch cached temporary credentials. [2]
  • Environment variables (AWS_ACCESS_KEY_ID / AWS_SECRET_ACCESS_KEY / AWS_SESSION_TOKEN). [1][3]
  • Assume-role / STS (aws sts assume-role or profiles that source a role). [3]
  • Credential_process / external providers (custom command that returns JSON credentials). [1][3]
  • On‑host providers: EC2 instance profile, ECS task credentials, or container metadata. [3]
  • Third‑party helpers (aws-vault, aws-okta, etc.) which wrap one of the above.

If you tell me which authentication flow you need (SSO, long‑term keys, role assumption, EC2/ECS, or a tooling helper), I’ll give the exact commands and example config.

Sources: [1] AWS CLI configure docs; [2] AWS CLI IAM Identity Center (SSO) docs; [3] AWS CLI config/credentials file and provider documentation.


Remove or correct the invalid aws login command.

aws login is not a valid AWS CLI command. AWS CLI supports specific credential methods like aws configure for static keys, aws sso login --profile <profile-name> for IAM Identity Center (SSO), or environment variables. The documentation link references AWS signin, which may not validate this specific command. Replace this with a supported credential method or remove Option D if it doesn't match an actual AWS credential approach Claude Code uses.

🤖 Prompt for AI Agents
In skills/working-with-claude-code/references/amazon-bedrock.md around lines 51
to 58, the example shows the invalid command `aws login`; replace or remove this
to avoid misinformation. Update Option D to present a valid credential method
such as using `aws configure` for static keys, `aws sso login --profile
<profile-name>` for IAM Identity Center, or explain using environment variables,
and adjust the accompanying link/text to reference the matching AWS CLI
credential documentation; if none of these apply to Claude Code, remove Option D
entirely.

#### Advanced credential configuration

Claude Code supports automatic credential refresh for AWS SSO and corporate identity providers. Add these settings to your Claude Code settings file (see [Settings](/en/docs/claude-code/settings) for file locations).
Claude Code supports automatic credential refresh for AWS SSO and corporate identity providers. Add these settings to your Claude Code settings file (see [Settings](/en/settings) for file locations).
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🔴 Critical

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
# Verify internal documentation links exist and sections/anchors are correct

# Check for Settings page
echo "=== Checking for settings.md page ==="
fd -t f "settings.md" 

# Check for VS Code page
echo "=== Checking for vs-code.md page ==="
fd -t f "vs-code.md" 

# Search for the specific section about third-party providers
echo "=== Searching for third-party providers section ==="
rg -n "using-third-party-providers|Using Third-Party Providers|Third-Party Providers" --type md -i

# Check the actual file structure under /en/ paths
echo "=== Listing markdown files under en directory ==="
find . -path "*/en/*.md" -type f | head -20

Repository: obra/superpowers-developing-for-claude-code

Length of output: 1740


🏁 Script executed:

#!/bin/bash
# Check the exact section heading and context around the third-party providers section
sed -n '45,60p' skills/working-with-claude-code/references/vs-code.md

Repository: obra/superpowers-developing-for-claude-code

Length of output: 690


🏁 Script executed:

#!/bin/bash
# Search for explicit anchor definitions (various markdown anchor formats)
echo "=== Checking for explicit anchors in vs-code.md ==="
rg -n "^\{#|<!-- ?#|id=|anchor=" skills/working-with-claude-code/references/vs-code.md

# Also check for the specific anchor text from the link
echo "=== Searching for the anchor text used in the link ==="
rg -n "using-third-party-providers-vertex-and-bedrock" skills/working-with-claude-code/references/vs-code.md

# Look at broader context around the section
echo "=== Broader context (lines 40-75) ==="
sed -n '40,75p' skills/working-with-claude-code/references/vs-code.md

Repository: obra/superpowers-developing-for-claude-code

Length of output: 1587


Fix incorrect anchor in VS Code link.

The link at line 113 references /en/vs-code#using-third-party-providers-vertex-and-bedrock, but this anchor does not exist in vs-code.md. The actual section heading is "#### Third-party providers and gateways" which generates the anchor #third-party-providers-and-gateways. Update the link to /en/vs-code#third-party-providers-and-gateways.

The files for /en/settings and /en/vs-code exist and are correct.

🤖 Prompt for AI Agents
In skills/working-with-claude-code/references/amazon-bedrock.md around line 69,
the VS Code link uses a nonexistent anchor
(/en/vs-code#using-third-party-providers-vertex-and-bedrock); update the link to
use the correct anchor /en/vs-code#third-party-providers-and-gateways so it
points to the "Third-party providers and gateways" section, ensuring the link
target is exactly /en/vs-code#third-party-providers-and-gateways.

Comment on lines +1 to +3
# null

# Claude Code Documentation Map
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟠 Major

Remove stray “# null” header (looks like a broken template variable).
This will confuse readers and any tooling parsing the map.

-# null
-
 # Claude Code Documentation Map
📝 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.

Suggested change
# null
# Claude Code Documentation Map
# Claude Code Documentation Map
🤖 Prompt for AI Agents
In skills/working-with-claude-code/references/claude_code_docs_map.md around
lines 1 to 3, remove the stray "# null" header (likely from a broken template
variable) so the file begins with the intended "# Claude Code Documentation Map"
header; ensure no blank or placeholder headers remain and save the file.

Comment on lines 7 to 12
<Note>
Claude Code GitHub Actions is built on top of the [Claude Code
SDK](/en/docs/claude-code/sdk), which enables programmatic integration of
SDK](https://docs.claude.com/en/docs/agent-sdk), which enables programmatic integration of
Claude Code into your applications. You can use the SDK to build custom
automation workflows beyond GitHub Actions.
</Note>
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟠 Major

🧩 Analysis chain

🌐 Web query:

Claude Code SDK documentation canonical URL 2025

💡 Result:

The canonical Claude Code SDK (renamed Claude Agent SDK) documentation URL in 2025 is:
https://docs.claude.com/en/api/agent-sdk. [1][2]

Notes: the SDK was renamed from "Claude Code SDK" to "Claude Agent SDK" and the docs/migration guide reflect this change. [1][2]

Sources:
[1] Agent SDK overview — Claude Docs.
[2] Migrate to Claude Agent SDK / Claude Code SDK — Claude Docs.


🌐 Web query:

docs.claude.com Claude Code SDK accessibility

💡 Result:

I searched docs.claude.com for information about the Claude Code SDK and accessibility. Findings:

  • The Claude Code SDK has been renamed/migrated to the Claude Agent SDK; migration guide and SDK docs (TS/Python) live on docs.claude.com. [1][2]
  • SDK docs cover features (permissions, hooks, auth, tools) and release notes but I did not find a dedicated accessibility (a11y) statement or section in the SDK docs pages I checked. [1][3][4]

If you want, I can (a) search the site more deeply for an accessibility statement/contact page, (b) open a specific docs page (e.g., the TypeScript or Python SDK reference) and look for accessibility info, or (c) draft an accessibility question you can send to Claude support. Which would you prefer?


🏁 Script executed:

# Check if the link is accessible and returns content
curl -s -o /dev/null -w "%{http_code}" "https://docs.claude.com/en/docs/agent-sdk"

Repository: obra/superpowers-developing-for-claude-code

Length of output: 90


Update SDK link to canonical URL and reflect SDK rename

The link should be updated to the current canonical URL https://docs.claude.com/en/api/agent-sdk (not /en/docs/agent-sdk). The Claude Code SDK was renamed to Claude Agent SDK in 2025, and the documentation location changed accordingly.

🤖 Prompt for AI Agents
In skills/working-with-claude-code/references/github-actions.md around lines 7
to 12, the SDK link and name are outdated: update the link from
https://docs.claude.com/en/docs/agent-sdk to the canonical URL
https://docs.claude.com/en/api/agent-sdk and change any mention of "Claude Code
SDK" to "Claude Agent SDK" to reflect the 2025 rename; ensure the anchor text
and URL are replaced consistently.

Comment on lines 11 to 13
<Note>
This integration is built on top of the [Claude Code CLI and SDK](/en/docs/claude-code/sdk), enabling programmatic use of Claude in your CI/CD jobs and custom automation workflows.
This integration is built on top of the [Claude Code CLI and SDK](https://docs.claude.com/en/docs/agent-sdk), enabling programmatic use of Claude in your CI/CD jobs and custom automation workflows.
</Note>
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟠 Major

🧩 Analysis chain

🌐 Web query:

Claude Code CLI SDK documentation current URL

💡 Result:

The current Claude Code CLI / SDK docs are at:

Sources: [1] Anthropic docs — Claude Code SDK; [2] Claude docs — Claude Code / migration notice.


🏁 Script executed:

# Check the file for all docs.claude.com and code.claude.com references
cat "skills/working-with-claude-code/references/gitlab-ci-cd.md" | grep -n "docs.claude.com\|code.claude.com" || echo "File not found or no matches"

Repository: obra/superpowers-developing-for-claude-code

Length of output: 517


🌐 Web query:

docs.claude.com agent-sdk endpoint authentication

💡 Result:

Short summary — how Agent SDK endpoint authentication works:

  • Primary (recommended): set ANTHROPIC_API_KEY with a Claude API key (environment variable). [1]
  • Alternate providers: configure AWS or Google credentials and set CLAUDE_CODE_USE_BEDROCK=1 or CLAUDE_CODE_USE_VERTEX=1 to authenticate via Amazon Bedrock or Google Vertex. [1]
  • OAuth / long‑lived tokens: some installs support a Claude OAuth token (e.g., CLAUDE_AGENT_OAUTH_TOKEN / CLAUDE_CODE OAuth flow) as an alternative to API keys (requires Anthropic approval/enablement). Many community containers require either ANTHROPIC_API_KEY or an OAuth token. [2][3]

If you want exact env var names, examples, or setup steps for Bedrock/Vertex or OAuth, tell me which one and I’ll paste the specific commands.

Sources: Claude Agent SDK docs and community README. [1] [2]


Update Claude Code documentation reference to the official public location

Line 12 links to https://docs.claude.com/en/docs/agent-sdk, which may require authentication. Update to the official public documentation at https://docs.anthropic.com/s/claude-code-sdk.

-  This integration is built on top of the [Claude Code CLI and SDK](https://docs.claude.com/en/docs/agent-sdk), enabling programmatic use of Claude in your CI/CD jobs and custom automation workflows.
+  This integration is built on top of the [Claude Code CLI and SDK](https://docs.anthropic.com/s/claude-code-sdk), enabling programmatic use of Claude in your CI/CD jobs and custom automation workflows.
📝 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.

Suggested change
<Note>
This integration is built on top of the [Claude Code CLI and SDK](/en/docs/claude-code/sdk), enabling programmatic use of Claude in your CI/CD jobs and custom automation workflows.
This integration is built on top of the [Claude Code CLI and SDK](https://docs.claude.com/en/docs/agent-sdk), enabling programmatic use of Claude in your CI/CD jobs and custom automation workflows.
</Note>
<Note>
This integration is built on top of the [Claude Code CLI and SDK](https://docs.anthropic.com/s/claude-code-sdk), enabling programmatic use of Claude in your CI/CD jobs and custom automation workflows.
</Note>
🤖 Prompt for AI Agents
In skills/working-with-claude-code/references/gitlab-ci-cd.md around lines 11 to
13, the documentation link points to a possibly restricted URL
(https://docs.claude.com/en/docs/agent-sdk); update the anchor to the public
Claude Code SDK docs by replacing that URL with
https://docs.anthropic.com/s/claude-code-sdk so the note links to the official
public documentation.

Comment on lines +27 to 29
* [Check system requirements](/en/setup#system-requirements)
* [Install Claude Code](/en/setup#installation)
* Login with Console account credentials
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟠 Major

Likely broken anchors: /en/setup#installation doesn’t match the setup page headings.
setup.md uses “Standard installation”, so these links probably 404-scroll.

-  * [Install Claude Code](/en/setup#installation)
+  * [Install Claude Code](/en/setup#standard-installation)
...
-3. Users can [install Claude Code](/en/setup#installation)
+3. Users can [install Claude Code](/en/setup#standard-installation)

Also applies to: 37-38

🤖 Prompt for AI Agents
In skills/working-with-claude-code/references/iam.md around lines 27-29 (and
also apply same fix to lines 37-38), the anchor /en/setup#installation is
incorrect because the setup page uses the heading "Standard installation";
update the link target to match the actual heading anchor (e.g.,
/en/setup#standard-installation or the canonical anchor produced by the site) or
change it to the correct relative URL/section name; verify the exact anchor in
setup.md and replace both occurrences so the links scroll correctly instead of
404-scrolling.

#### General Settings

* **Claude command**: Specify a custom command to run Claude (e.g., `claude`, `/usr/local/bin/claude`, or `npx @anthropic/claude`)
* **Claude command**: Specify a custom command to run Claude (for example, `claude`, `/usr/local/bin/claude`, or `npx @anthropic/claude`)
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🔴 Critical

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
# Search for both package name patterns in the repository
echo "=== Searching for 'npx @anthropic/claude' ==="
rg -n --fixed-strings 'npx @anthropic/claude' -S .

echo ""
echo "=== Searching for '@anthropic-ai/claude-code' ==="
rg -n --fixed-strings '@anthropic-ai/claude-code' -S .

echo ""
echo "=== Searching for package references in package.json files ==="
fd -e package.json -x cat {} +

echo ""
echo "=== Checking for claude-related documentation ==="
fd -t f \( -name "*install*" -o -name "*setup*" -o -name "*README*" \) | head -20

Repository: obra/superpowers-developing-for-claude-code

Length of output: 3482


🏁 Script executed:

#!/bin/bash
# More focused search for claude package references
echo "=== All occurrences of 'claude' in .md files ==="
rg -n 'claude' --type md | grep -i 'npm\|npx\|@anthropic'

echo ""
echo "=== Checking the specific file mentioned ==="
cat -n skills/working-with-claude-code/references/jetbrains.md | grep -A 3 -B 3 "Claude command"

Repository: obra/superpowers-developing-for-claude-code

Length of output: 6256


Fix incorrect npm package name in example command.

The example at line 71 uses npx @anthropic/claude, but the correct package name throughout the documentation (and confirmed by line 134 in the same file) is @anthropic-ai/claude-code. Update the example to npx @anthropic-ai/claude-code to match the installation documentation and prevent user errors.

🤖 Prompt for AI Agents
In skills/working-with-claude-code/references/jetbrains.md around line 71 the
example command uses the incorrect npm package name `npx @anthropic/claude`;
update the example to use the correct package name `npx
@anthropic-ai/claude-code` so it matches the rest of the documentation and the
installation instructions.

Comment on lines +158 to +175
### Resetting configuration

To reset Claude Code to default settings, you can remove the configuration files:

```bash theme={null}
# Reset all user settings and state
rm ~/.claude.json
rm -rf ~/.claude/

# Reset project-specific settings
rm -rf .claude/
rm .mcp.json
```

<Warning>
This will remove all your settings, allowed tools, MCP server configurations, and session history.
</Warning>

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟠 Major

Reset commands are destructive: add -f safeguards + mention Windows/PowerShell equivalent.

As written, rm .mcp.json will error if the file doesn’t exist (and rm ~/.claude.json too). Suggest using rm -f and rm -rf consistently, and consider adding a short PowerShell snippet or a note that these commands are for bash/zsh/WSL.

- rm ~/.claude.json
- rm -rf ~/.claude/
+ rm -f ~/.claude.json
+ rm -rf ~/.claude/

- rm -rf .claude/
- rm .mcp.json
+ rm -rf .claude/
+ rm -f .mcp.json
📝 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.

Suggested change
### Resetting configuration
To reset Claude Code to default settings, you can remove the configuration files:
```bash theme={null}
# Reset all user settings and state
rm ~/.claude.json
rm -rf ~/.claude/
# Reset project-specific settings
rm -rf .claude/
rm .mcp.json
```
<Warning>
This will remove all your settings, allowed tools, MCP server configurations, and session history.
</Warning>
### Resetting configuration
To reset Claude Code to default settings, you can remove the configuration files:
🤖 Prompt for AI Agents
In skills/working-with-claude-code/references/troubleshooting.md around lines
158 to 175, the reset commands are unsafe on POSIX shells and lack
Windows/PowerShell equivalents; update the examples to use force flags (rm -f
for files, rm -rf for directories) so removal is non-erroring if targets are
absent, ensure directory removals use -rf consistently, and add a short note or
a PowerShell snippet showing the equivalent commands (e.g., Remove-Item -Recurse
-Force) or explicitly state the bash/zsh/WSL context for the shown commands.

## Overview

This skill provides complete, authoritative documentation for Claude Code directly from docs.claude.com. Instead of guessing about configuration paths, API structures, or feature capabilities, read the official docs stored in this skill's references directory.
This skill provides complete, authoritative documentation for Claude Code directly from code.claude.com. Instead of guessing about configuration paths, API structures, or feature capabilities, read the official docs stored in this skill's references directory.
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🔴 Critical

🧩 Analysis chain

🌐 Web query:

official Claude Code documentation domain site:anthropic.com

💡 Result:

Here are the official Claude Code documentation pages on Anthropic’s site:

  • Claude Code overview (official docs): docs.anthropic.com/en/docs/claude-code/overview. [1]
  • Claude Code SDK / Quickstart: docs.anthropic.com/s/claude-code-sdk. [2]
  • Product page / features: anthropic.com/claude-code. [3]

Sources: [1] [2] [3].


🌐 Web query:

code.claude.com Claude Code docs

💡 Result:

You can find the official Claude Code docs on Anthropic’s site:

(There are also third‑party mirrors/summaries, but use the Anthropic docs above for the authoritative reference.) [3]


Update domain reference: code.claude.com is not the official Claude Code documentation source.

The official Claude Code documentation is hosted at docs.anthropic.com/en/docs/claude-code/overview, not code.claude.com. SKILL.md should reference the authoritative Anthropic documentation domain (docs.anthropic.com) instead of the unverified code.claude.com domain to ensure accuracy and stability for documentation consumers.

🤖 Prompt for AI Agents
In skills/working-with-claude-code/SKILL.md around line 10, the domain reference
points to code.claude.com which is not the authoritative source; update the
reference to the official Claude Code docs at
docs.anthropic.com/en/docs/claude-code/overview, replacing any occurrences of
code.claude.com with the docs.anthropic.com URL and ensure any link text and
context reflect the authoritative Anthropic documentation domain.

Comment on lines 46 to 94
```
references/
├── overview.md # Claude Code introduction
├── quickstart.md # Getting started guide
├── setup.md # Installation and setup
├── plugins.md # Plugin development
├── plugins-reference.md # Plugin API reference
├── plugin-marketplaces.md # Plugin marketplaces
├── skills.md # Skill creation
├── mcp.md # MCP server integration
├── hooks.md # Hooks overview
├── hooks-guide.md # Hooks implementation guide
├── slash-commands.md # Slash command reference
├── sub-agents.md # Subagent usage
├── settings.md # Configuration reference
├── cli-reference.md # CLI command reference
├── common-workflows.md # Common usage patterns
├── interactive-mode.md # Interactive mode guide
├── headless.md # Headless mode guide
├── output-styles.md # Output customization
├── statusline.md # Status line configuration
├── memory.md # Memory and context management
├── checkpointing.md # Checkpointing feature
├── analytics.md # Usage analytics
├── costs.md # Cost tracking
├── monitoring-usage.md # Usage monitoring
├── data-usage.md # Data usage policies
├── security.md # Security features
├── iam.md # IAM integration
├── network-config.md # Network configuration
├── terminal-config.md # Terminal configuration
├── model-config.md # Model configuration
├── llm-gateway.md # LLM gateway setup
├── amazon-bedrock.md # AWS Bedrock integration
├── google-vertex-ai.md # Google Vertex AI integration
├── vs-code.md # VS Code integration
├── jetbrains.md # JetBrains integration
├── devcontainer.md # Dev container support
├── github-actions.md # GitHub Actions integration
├── gitlab-ci-cd.md # GitLab CI/CD integration
├── quickstart.md # Getting started guide
├── setup.md # Installation and setup
├── common-workflows.md # Common usage patterns
├── claude-code-on-the-web.md # Web-based Claude Code
├── desktop.md # Desktop application
├── plugins.md # Plugin development
├── plugins-reference.md # Plugin API reference
├── plugin-marketplaces.md # Plugin marketplaces
├── skills.md # Skill creation
├── mcp.md # MCP server integration
├── hooks.md # Hooks overview
├── hooks-guide.md # Hooks implementation guide
├── slash-commands.md # Slash command reference
├── sub-agents.md # Subagent usage
├── settings.md # Configuration reference
├── cli-reference.md # CLI command reference
├── interactive-mode.md # Interactive mode guide
├── headless.md # Headless mode guide
├── output-styles.md # Output customization
├── statusline.md # Status line configuration
├── memory.md # Memory and context management
├── checkpointing.md # Checkpointing feature
├── sandboxing.md # Sandboxing and security isolation
├── analytics.md # Usage analytics
├── costs.md # Cost tracking
├── monitoring-usage.md # Usage monitoring
├── data-usage.md # Data usage policies
├── security.md # Security features
├── iam.md # IAM integration
├── network-config.md # Network configuration
├── terminal-config.md # Terminal configuration
├── model-config.md # Model configuration
├── llm-gateway.md # LLM gateway setup
├── amazon-bedrock.md # AWS Bedrock integration
├── google-vertex-ai.md # Google Vertex AI integration
├── microsoft-foundry.md # Microsoft Azure AI Foundry
├── vs-code.md # VS Code integration
├── jetbrains.md # JetBrains integration
├── devcontainer.md # Dev container support
├── github-actions.md # GitHub Actions integration
├── gitlab-ci-cd.md # GitLab CI/CD integration
├── slack.md # Slack integration
├── third-party-integrations.md # Other integrations
├── legal-and-compliance.md # Legal information
├── troubleshooting.md # Troubleshooting guide
└── migration-guide.md # Migration guide
├── legal-and-compliance.md # Legal information
└── troubleshooting.md # Troubleshooting guide
```
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟠 Major

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
set -euo pipefail

# First, check if the SKILL.md file exists
if [ ! -f "skills/working-with-claude-code/SKILL.md" ]; then
  echo "File not found: skills/working-with-claude-code/SKILL.md"
  exit 1
fi

# Run the provided verification script
python - <<'PY'
import re, pathlib, sys
p = pathlib.Path("skills/working-with-claude-code/SKILL.md")
text = p.read_text(encoding="utf-8")
# pull markdown filenames from the tree block lines
names = sorted(set(re.findall(r'`?([a-z0-9][a-z0-9\-]*\.md)`?', text)))
base = pathlib.Path("skills/working-with-claude-code/references")
missing = [n for n in names if not (base/n).exists()]
if missing:
    print("Missing reference files:")
    for n in missing: print(" -", n)
    sys.exit(1)
print(f"OK: {len(names)} referenced .md files exist under references/")
PY

Repository: obra/superpowers-developing-for-claude-code

Length of output: 173


🏁 Script executed:

# Also, let's list the actual contents of the references directory to see what exists
if [ -d "skills/working-with-claude-code/references" ]; then
  echo "=== Actual files in references/ directory ==="
  ls -1 "skills/working-with-claude-code/references" | grep "\.md$" | sort
else
  echo "references/ directory does not exist"
fi

Repository: obra/superpowers-developing-for-claude-code

Length of output: 884


Update references/ tree to include missing files: claude_code_docs_map.md and migration-guide.md.

The tree block in SKILL.md (lines 46-94) lists 45 files, but the actual references/ directory contains 48 files. The two missing entries are claude_code_docs_map.md and migration-guide.md. To prevent future drift, consider auto-generating this tree from the filesystem (e.g., via scripts/update_docs.js) rather than maintaining it manually.

🧰 Tools
🪛 markdownlint-cli2 (0.18.1)

46-46: Fenced code blocks should have a language specified

(MD040, fenced-code-language)

🤖 Prompt for AI Agents
In skills/working-with-claude-code/SKILL.md around lines 46 to 94, the
references/ tree is missing two files (claude_code_docs_map.md and
migration-guide.md); update the listed tree to include these two entries in the
appropriate alphabetical/structural positions, and to prevent future drift add
or update a docs generation step (e.g., scripts/update_docs.js) that reads the
references/ directory and regenerates the tree block automatically as part of
the docs build or a pre-commit hook.

Copy link

@coderabbitai coderabbitai bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actionable comments posted: 0

🧹 Nitpick comments (4)
docs/upstream-improvements-context.md (3)

28-32: Add language identifiers to fenced code blocks for syntax highlighting.

The document contains multiple code blocks without language specifications. Adding these improves readability and enables proper syntax highlighting when the file is rendered.

Example fix:

<!-- Before -->
```
❌ Error: HTTP 401: Unauthorized
🚀 Claude Code Documentation Updater
```

<!-- After -->
```text
❌ Error: HTTP 401: Unauthorized
🚀 Claude Code Documentation Updater
```

<!-- For bash commands -->
```bash
WebFetch(url='https://docs.claude.com/llms.txt')
# Result: Request failed with status code 401
```

Affected lines include: 28, 35, 56, 231, 464, 487, 529, 655. Use bash, javascript, json, text, or other appropriate language identifiers based on content.

Also applies to: 35-37, 56-66, 231-247, 464-484, 487-489, 529-537, 655-661


341-341: Convert emphasis-based labels to proper heading hierarchy.

Lines 341 and 346 use bold emphasis (**text**) for section labels that function as headings. Consider upgrading to proper markdown headings for better document structure:

<!-- Line 341 - Before -->
**1. update_docs.js**

<!-- Line 341 - After -->
### 1. update_docs.js

This improves outline navigation, accessibility, and consistency with the rest of the document's heading hierarchy.

Also applies to: 346-346


385-385: Wrap bare URLs in markdown link syntax.

Lines with bare URLs (e.g., line 385: https://github.com/obra/episodic-memory) should be wrapped in markdown link syntax for consistency and better rendering. Example:

<!-- Before -->
URL: https://github.com/obra/episodic-memory

<!-- After -->
Repository: [https://github.com/obra/episodic-memory](https://github.com/obra/episodic-memory)
<!-- or -->
Repository: [obra/episodic-memory](https://github.com/obra/episodic-memory)

Affected lines: 385, 581, 950, 951, 955, 956.

Also applies to: 581-581, 950-951, 955-956

docs/working-with-claude-code-update-fix.md (1)

6-6: Fix markdown linting violations for better documentation consistency.

The document has several formatting issues flagged by markdownlint:

  • Lines 6, 27, 131: Fenced code blocks should have language identifiers
  • Lines 156, 179, 184: Bare URLs should be wrapped in angle brackets or markdown links
  • Lines 159, 164: Use proper heading syntax (###) instead of bold emphasis for section titles

Apply this diff to resolve all linting violations:

-```
-❌ Error: HTTP 401: Unauthorized
-📥 Fetching llms.txt...
-```
+```plaintext
+❌ Error: HTTP 401: Unauthorized
+📥 Fetching llms.txt...
+```

-```
-~/.claude/plugins/cache/superpowers-marketplace/superpowers-developing-for-claude-code/0.3.1/skills/working-with-claude-code/scripts/update_docs.js
-```
+```plaintext
+~/.claude/plugins/cache/superpowers-marketplace/superpowers-developing-for-claude-code/0.3.1/skills/working-with-claude-code/scripts/update_docs.js
+```

-```
-Claude Code Documentation Updater
-
-Found 43 Claude Code documentation pages
-
-Downloading documentation...
-  Fetching overview.md...
-  [... 42 more files ...]
-
-Documentation update complete!
-   43 files downloaded successfully
-
-Documentation saved to: [path]/references
-```
+```plaintext
+Claude Code Documentation Updater
+
+Found 43 Claude Code documentation pages
+
+Downloading documentation...
+  Fetching overview.md...
+  [... 42 more files ...]
+
+Documentation update complete!
+   43 files downloaded successfully
+
+Documentation saved to: [path]/references
+```

-**Option 1: Contribute upstream** (Best)
-- Open an issue/PR at: https://github.com/obra/superpowers-developing-for-claude-code
+### Option 1: Contribute Upstream (Best)
+
+- Open an issue/PR at: <https://github.com/obra/superpowers-developing-for-claude-code>

-**Option 2: Document in episodic-memory**
+### Option 2: Document in Episodic Memory
+
 - This conversation is now indexed and searchable
 - Search: `episodic-memory search "update docs script 401"`
 - Will find this fix in future sessions

-**Option 3: Keep this documentation**
+### Option 3: Keep This Documentation
+
 - Commit this file to your repo
 - Reference it when the fix needs to be reapplied
 - Consider adding to your BEADS workflow

-## Related Issues
-
-- Claude Code documentation moved from docs.claude.com → code.claude.com (Nov 2025)
-- llms.txt endpoint now requires authentication
-- See: https://code.claude.com/docs/en/overview for current documentation
+## Related Issues
+
+- Claude Code documentation moved from docs.claude.com → code.claude.com (Nov 2025)
+- llms.txt endpoint now requires authentication
+- See: <https://code.claude.com/docs/en/overview> for current documentation

 ## Next Steps
 
 1. ✅ Document the fix (this file)
-2. ⬜ Open issue at https://github.com/obra/superpowers-developing-for-claude-code
+2. ⬜ Open issue at <https://github.com/obra/superpowers-developing-for-claude-code>

Also applies to: 27-27, 131-131, 156-156, 159-159, 164-164, 179-179, 184-184

📜 Review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 5be2f6f and a3cc3db.

📒 Files selected for processing (2)
  • docs/upstream-improvements-context.md (1 hunks)
  • docs/working-with-claude-code-update-fix.md (1 hunks)
🧰 Additional context used
🪛 markdownlint-cli2 (0.18.1)
docs/working-with-claude-code-update-fix.md

6-6: Fenced code blocks should have a language specified

(MD040, fenced-code-language)


27-27: Fenced code blocks should have a language specified

(MD040, fenced-code-language)


131-131: Fenced code blocks should have a language specified

(MD040, fenced-code-language)


156-156: Bare URL used

(MD034, no-bare-urls)


159-159: Emphasis used instead of a heading

(MD036, no-emphasis-as-heading)


164-164: Emphasis used instead of a heading

(MD036, no-emphasis-as-heading)


179-179: Bare URL used

(MD034, no-bare-urls)


184-184: Bare URL used

(MD034, no-bare-urls)

docs/upstream-improvements-context.md

28-28: Fenced code blocks should have a language specified

(MD040, fenced-code-language)


35-35: Fenced code blocks should have a language specified

(MD040, fenced-code-language)


56-56: Fenced code blocks should have a language specified

(MD040, fenced-code-language)


231-231: Fenced code blocks should have a language specified

(MD040, fenced-code-language)


341-341: Emphasis used instead of a heading

(MD036, no-emphasis-as-heading)


346-346: Emphasis used instead of a heading

(MD036, no-emphasis-as-heading)


385-385: Bare URL used

(MD034, no-bare-urls)


464-464: Fenced code blocks should have a language specified

(MD040, fenced-code-language)


487-487: Fenced code blocks should have a language specified

(MD040, fenced-code-language)


529-529: Fenced code blocks should have a language specified

(MD040, fenced-code-language)


581-581: Bare URL used

(MD034, no-bare-urls)


655-655: Fenced code blocks should have a language specified

(MD040, fenced-code-language)


950-950: Bare URL used

(MD034, no-bare-urls)


951-951: Bare URL used

(MD034, no-bare-urls)


955-955: Bare URL used

(MD034, no-bare-urls)


956-956: Bare URL used

(MD034, no-bare-urls)

🔇 Additional comments (1)
docs/upstream-improvements-context.md (1)

1-989: LGTM — Comprehensive context document with excellent detail and structure.

This document provides thorough context for upstream contributions, including root cause analysis, verified solutions, testing results, and clear recommendations. The writing is clear, well-organized, and actionable. The investigation work is methodical and well-documented, which will be valuable for the maintainers reviewing PRs based on this context.

The only improvements needed are minor markdown formatting polishes (language specifiers, heading hierarchy, link syntax), which do not affect the substantive quality or utility of the document.

The working-with-claude-code-update-fix.md file is completely redundant with upstream-improvements-context.md, which contains all the same information plus additional episodic-memory setup documentation.
- Update 33 reference files with latest content from code.claude.com
- Add chrome.md for Chrome extension documentation
- Add discover-plugins.md for plugin discovery documentation
- Update SKILL.md with new file references and quick reference entries
Copy link

@coderabbitai coderabbitai bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actionable comments posted: 4

🤖 Fix all issues with AI agents
In `@skills/working-with-claude-code/references/plugin-marketplaces.md`:
- Around line 562-564: Replace the bare URL in the sentence starting "To find
navigation and other pages in this documentation, fetch the llms.txt file at:
https://code.claude.com/docs/llms.txt" with a proper Markdown link (e.g., use
[https://code.claude.com/docs/llms.txt](https://code.claude.com/docs/llms.txt)
or a descriptive link label) to satisfy markdownlint MD034; update the line in
plugin-marketplaces.md where that sentence appears so the URL is not bare.

In `@skills/working-with-claude-code/references/plugins-reference.md`:
- Around line 746-748: The bare URL in the sentence "To find navigation and
other pages in this documentation, fetch the llms.txt file at:
https://code.claude.com/docs/llms.txt" triggers MD034; change it to a non-bare
URL by either converting it into a Markdown link (e.g., use "fetch the llms.txt
file at
[https://code.claude.com/docs/llms.txt](https://code.claude.com/docs/llms.txt)"
or "fetch the llms.txt file at
[llms.txt](https://code.claude.com/docs/llms.txt)") or by wrapping the URL in
angle brackets (<https://code.claude.com/docs/llms.txt>); update the text where
that exact sentence/URL appears in plugins-reference.md.

In `@skills/working-with-claude-code/references/plugins.md`:
- Around line 204-212: The fenced code block showing the directory tree in
plugins.md lacks a language tag; update the opening fence for the directory tree
(the triple-backtick block that contains "my-plugin/ ├── .claude-plugin/ ...")
to include a language tag such as ```text so markdownlint stops flagging it,
keeping the content and closing ``` unchanged; locate the block in
skills/working-with-claude-code/references/plugins.md and modify the opening
fence accordingly.

In `@skills/working-with-claude-code/references/statusline.md`:
- Around line 268-270: Wrap the bare URL in the footer text "To find navigation
and other pages in this documentation, fetch the llms.txt file at:
https://code.claude.com/docs/llms.txt" so it satisfies markdownlint MD034;
either enclose the URL in angle brackets
(<https://code.claude.com/docs/llms.txt>) or convert it to link text (e.g.,
"fetch the llms.txt file at
[https://code.claude.com/docs/llms.txt](https://code.claude.com/docs/llms.txt)"
or "fetch the [llms.txt](https://code.claude.com/docs/llms.txt) file") and
update that line accordingly.
♻️ Duplicate comments (8)
skills/working-with-claude-code/references/gitlab-ci-cd.md (2)

11-13: Update the SDK documentation link to the official Anthropic domain.

Line 12 still references https://docs.claude.com/en/docs/agent-sdk, which may require authentication. The official public documentation is at https://docs.anthropic.com/s/claude-code-sdk.

📝 Proposed fix
-  This integration is built on top of the [Claude Code CLI and SDK](https://docs.claude.com/en/docs/agent-sdk), enabling programmatic use of Claude in your CI/CD jobs and custom automation workflows.
+  This integration is built on top of the [Claude Code CLI and SDK](https://docs.anthropic.com/s/claude-code-sdk), enabling programmatic use of Claude in your CI/CD jobs and custom automation workflows.

404-408: Update the pricing link to the canonical Anthropic documentation domain.

Line 407 references https://docs.claude.com/en/docs/about-claude/pricing, but the official pricing documentation is at https://docs.anthropic.com/en/docs/about-claude/pricing (or https://claude.com/pricing for the public-facing page).

📝 Proposed fix
-  * See [Anthropic pricing](https://docs.claude.com/en/docs/about-claude/pricing) for details
+  * See [Anthropic pricing](https://docs.anthropic.com/en/docs/about-claude/pricing) for details
skills/working-with-claude-code/references/claude_code_docs_map.md (1)

1-3: Remove the stray # null header.

Lines 1-2 contain a # null header that appears to be a broken template variable or placeholder. This should be removed so the file begins with the intended # Claude Code Documentation Map header.

📝 Proposed fix
-# null
-
 # Claude Code Documentation Map
skills/working-with-claude-code/references/troubleshooting.md (1)

193-200: Make reset commands non‑erroring (rm -f) and clarify shell context.
Line 195 and Line 200 use rm without -f; this can fail if files are missing. Also consider a short PowerShell equivalent note since Windows users are targeted here.

✏️ Suggested fix
-rm ~/.claude.json
+rm -f ~/.claude.json
 rm -rf ~/.claude/
 
 # Reset project-specific settings
 rm -rf .claude/
-rm .mcp.json
+rm -f .mcp.json
skills/working-with-claude-code/references/iam.md (1)

43-45: Likely broken setup anchor (verify target heading).

The /en/setup#installation anchor was previously flagged as mismatched to the setup page headings. Please confirm the correct anchor in setup.md and update both occurrences accordingly.

Run this to confirm the actual heading IDs in skills/working-with-claude-code/references/setup.md:

#!/bin/bash
# Find headings around installation in setup.md
rg -n '^##|^###' skills/working-with-claude-code/references/setup.md

Also applies to: 53-53

skills/working-with-claude-code/references/hooks.md (1)

246-255: Remove or correct the "Supported hook events" section - documentation still contradicts itself.

This contradiction was flagged in a previous review but remains unresolved. Line 193 explicitly states that prompt-based hooks are "currently only supported for Stop and SubagentStop hooks", yet this section lists additional events (UserPromptSubmit, PreToolUse, PermissionRequest) as supported.

Either:

  1. Remove this entire section, or
  2. Limit it to only Stop and SubagentStop with a note clarifying that other events require command-type hooks
skills/working-with-claude-code/SKILL.md (2)

10-10: Confirm the authoritative documentation domain before labeling it “official.”
This repeats a prior concern: if code.claude.com is not the authoritative Claude Code docs source, SKILL.md should not describe it as such. Please re‑verify the official source and update wording/links accordingly.

What is the official Claude Code documentation domain (authoritative source) as of 2026?

Also applies to: 135-148


47-96: Keep the references/ tree in sync with actual files.
This matches a previous comment: the tree omits claude_code_docs_map.md and migration-guide.md. Add them (or auto‑generate the tree) to prevent drift.

🧹 Nitpick comments (15)
skills/working-with-claude-code/references/statusline.md (1)

236-248: Include output tokens in manual context usage calculation.

The current formula only sums input and cache tokens, which undercounts usage when output tokens are non-zero. Consider including .output_tokens to match actual context consumption.

♻️ Proposed fix
-    CURRENT_TOKENS=$(echo "$USAGE" | jq '.input_tokens + .cache_creation_input_tokens + .cache_read_input_tokens')
+    CURRENT_TOKENS=$(echo "$USAGE" | jq '.input_tokens + .output_tokens + .cache_creation_input_tokens + .cache_read_input_tokens')
skills/working-with-claude-code/references/data-usage.md (1)

46-48: Hyphenate “first‑party” for consistency.
Line 48 uses “first party users”; elsewhere this is typically hyphenated as a compound modifier.

✏️ Proposed edit
-For all first party users, you can learn more about what data is logged for [local Claude Code](`#local-claude-code-data-flow-and-dependencies`) and [remote Claude Code](`#cloud-execution-data-flow-and-dependencies`).
+For all first-party users, you can learn more about what data is logged for [local Claude Code](`#local-claude-code-data-flow-and-dependencies`) and [remote Claude Code](`#cloud-execution-data-flow-and-dependencies`).
skills/working-with-claude-code/references/mcp.md (1)

5-10: Ensure React hooks are available in MDX runtime.

useState and useEffect are referenced without imports. If this MDX environment doesn’t auto-inject React hooks, the page will fail to render.

✅ Suggested fix (if imports required)
+import { useEffect, useState } from "react";
skills/working-with-claude-code/references/claude-code-on-the-web.md (4)

64-66: Add language identifier to fenced code block.

The code block showing user input would benefit from a language identifier for proper syntax highlighting.

Proposed fix
-```
-& Fix the authentication bug in src/auth/login.ts
-```
+```bash
+& Fix the authentication bug in src/auth/login.ts
+```

86-88: Add language identifier to fenced code block.

The code block showing user input would benefit from a language identifier for proper syntax highlighting.

Proposed fix
-```
-& Execute the migration plan we discussed
-```
+```bash
+& Execute the migration plan we discussed
+```

193-196: Add language identifier for .env format code block.

The code block shows .env format examples and should specify a language identifier.

Proposed fix
-  ```
-  API_KEY=your_api_key
-  DEBUG=true
-  ```
+  ```env
+  API_KEY=your_api_key
+  DEBUG=true
+  ```

552-552: Use markdown link format for better accessibility.

The bare URL in the navigation hint should use markdown link format for improved accessibility and consistency with markdown best practices.

Proposed fix
-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt)
skills/working-with-claude-code/references/sandboxing.md (1)

65-67: Add language identifier to fenced code block.

The code block showing the /sandbox command should specify a language identifier.

Proposed fix
-```
-> /sandbox
-```
+```bash
+> /sandbox
+```
skills/working-with-claude-code/references/desktop.md (3)

53-58: Add language identifier to fenced code block.

The .worktreeinclude patterns example should specify a language identifier such as gitignore or text.

Proposed fix
-```
-.env
-.env.local
-.env.*
-**/.claude/settings.local.json
-```
+```gitignore
+.env
+.env.local
+.env.*
+**/.claude/settings.local.json
+```

93-101: Add language identifier for .env format code block.

The code block shows .env format examples and should specify the env language identifier.

Proposed fix
-  ```
-  API_KEY=your_api_key
-  DEBUG=true
-
-  # Multiline values - wrap in quotes
-  CERT="-----BEGIN CERT-----
-  MIIE...
-  -----END CERT-----"
-  ```
+  ```env
+  API_KEY=your_api_key
+  DEBUG=true
+
+  # Multiline values - wrap in quotes
+  CERT="-----BEGIN CERT-----
+  MIIE...
+  -----END CERT-----"
+  ```

119-119: Use markdown link format for better accessibility.

The bare URL in the navigation hint should use markdown link format for improved accessibility and consistency.

Proposed fix
-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt)
skills/working-with-claude-code/references/security.md (2)

90-90: Use descriptive link text.

The link text "here" is not descriptive. Consider using text that describes the destination.

Proposed fix
-See [here](/en/vs-code#security) for more information on the security of running Claude Code in an IDE.
+See [VS Code Security](/en/vs-code#security) for more information on the security of running Claude Code in an IDE.

141-141: Use markdown link format for better accessibility.

The bare URL in the navigation hint should use markdown link format for improved accessibility and consistency.

Proposed fix
-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt)
skills/working-with-claude-code/references/cli-reference.md (1)

154-154: Use markdown link format for better accessibility.

The bare URL in the navigation hint should use markdown link format for improved accessibility and consistency with other documentation files.

Proposed fix
-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the [llms.txt file](https://code.claude.com/docs/llms.txt)
skills/working-with-claude-code/references/hooks.md (1)

717-717: Consider removing redundant "back" from "return output back".

The phrase "return output back" is redundant. Consider revising to either "return output to Claude Code" or "send output back to Claude Code" for clearer writing.

Comment on lines +562 to +564
---

> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt No newline at end of file
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟡 Minor

Fix MD034 bare URL lint.

The bare https://code.claude.com/docs/llms.txt will trip markdownlint (MD034). Wrap it in a proper link.

✅ Suggested fix
-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the llms.txt file at: [code.claude.com/docs/llms.txt](https://code.claude.com/docs/llms.txt)
📝 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.

Suggested change
---
> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
---
> To find navigation and other pages in this documentation, fetch the llms.txt file at: [code.claude.com/docs/llms.txt](https://code.claude.com/docs/llms.txt)
🧰 Tools
🪛 markdownlint-cli2 (0.18.1)

564-564: Bare URL used

(MD034, no-bare-urls)

🤖 Prompt for AI Agents
In `@skills/working-with-claude-code/references/plugin-marketplaces.md` around
lines 562 - 564, Replace the bare URL in the sentence starting "To find
navigation and other pages in this documentation, fetch the llms.txt file at:
https://code.claude.com/docs/llms.txt" with a proper Markdown link (e.g., use
[https://code.claude.com/docs/llms.txt](https://code.claude.com/docs/llms.txt)
or a descriptive link label) to satisfy markdownlint MD034; update the line in
plugin-marketplaces.md where that sentence appears so the URL is not bare.

Comment on lines +746 to +748
---

> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt No newline at end of file
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟡 Minor

Fix MD034 bare URL lint.

The bare https://code.claude.com/docs/llms.txt will fail MD034 if enforced.

✅ Suggested fix
-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the llms.txt file at: [code.claude.com/docs/llms.txt](https://code.claude.com/docs/llms.txt)
📝 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.

Suggested change
---
> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
---
> To find navigation and other pages in this documentation, fetch the llms.txt file at: [code.claude.com/docs/llms.txt](https://code.claude.com/docs/llms.txt)
🤖 Prompt for AI Agents
In `@skills/working-with-claude-code/references/plugins-reference.md` around lines
746 - 748, The bare URL in the sentence "To find navigation and other pages in
this documentation, fetch the llms.txt file at:
https://code.claude.com/docs/llms.txt" triggers MD034; change it to a non-bare
URL by either converting it into a Markdown link (e.g., use "fetch the llms.txt
file at
[https://code.claude.com/docs/llms.txt](https://code.claude.com/docs/llms.txt)"
or "fetch the llms.txt file at
[llms.txt](https://code.claude.com/docs/llms.txt)") or by wrapping the URL in
angle brackets (<https://code.claude.com/docs/llms.txt>); update the text where
that exact sentence/URL appears in plugins-reference.md.

Comment on lines 204 to 212
```

```shell Browse available plugins theme={null}
/plugin
my-plugin/
├── .claude-plugin/
│ └── plugin.json
└── skills/
└── code-review/
└── SKILL.md
```

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟡 Minor

Add a language tag to the directory tree fence.
Line 204 opens a fenced block without a language; markdownlint flags this.

✏️ Suggested fix
-```
+```text
 my-plugin/
 ├── .claude-plugin/
 │   └── plugin.json
 └── skills/
     └── code-review/
         └── SKILL.md
</details>

<!-- suggestion_start -->

<details>
<summary>📝 Committable suggestion</summary>

> ‼️ **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.

```suggestion

🧰 Tools
🪛 markdownlint-cli2 (0.18.1)

204-204: Fenced code blocks should have a language specified

(MD040, fenced-code-language)


204-204: Code block style
Expected: indented; Actual: fenced

(MD046, code-block-style)

🤖 Prompt for AI Agents
In `@skills/working-with-claude-code/references/plugins.md` around lines 204 -
212, The fenced code block showing the directory tree in plugins.md lacks a
language tag; update the opening fence for the directory tree (the
triple-backtick block that contains "my-plugin/ ├── .claude-plugin/ ...") to
include a language tag such as ```text so markdownlint stops flagging it,
keeping the content and closing ``` unchanged; locate the block in
skills/working-with-claude-code/references/plugins.md and modify the opening
fence accordingly.

Comment on lines +268 to +270
---

> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt No newline at end of file
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟡 Minor

Fix bare URL to satisfy markdownlint (MD034).

The footer uses a bare URL; wrap it in angle brackets or use link text.

✅ Suggested fix
-> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
+> To find navigation and other pages in this documentation, fetch the llms.txt file at: <https://code.claude.com/docs/llms.txt>
📝 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.

Suggested change
---
> To find navigation and other pages in this documentation, fetch the llms.txt file at: https://code.claude.com/docs/llms.txt
---
> To find navigation and other pages in this documentation, fetch the llms.txt file at: <https://code.claude.com/docs/llms.txt>
🧰 Tools
🪛 markdownlint-cli2 (0.18.1)

270-270: Bare URL used

(MD034, no-bare-urls)

🤖 Prompt for AI Agents
In `@skills/working-with-claude-code/references/statusline.md` around lines 268 -
270, Wrap the bare URL in the footer text "To find navigation and other pages in
this documentation, fetch the llms.txt file at:
https://code.claude.com/docs/llms.txt" so it satisfies markdownlint MD034;
either enclose the URL in angle brackets
(<https://code.claude.com/docs/llms.txt>) or convert it to link text (e.g.,
"fetch the llms.txt file at
[https://code.claude.com/docs/llms.txt](https://code.claude.com/docs/llms.txt)"
or "fetch the [llms.txt](https://code.claude.com/docs/llms.txt) file") and
update that line accordingly.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant