bitwarden · theMickster · Dec 24, 2025 · Dec 23, 2025
@@ -93,12 +93,26 @@ Install via [claude.ai/download](https://claude.ai/download) or Homebrew `brew i
 - Configure your workspace preferences
 - Enable MCP server connections in Settings → Developer → MCP Servers
 
-## MCP servers
+## MCP servers and extensions
 
 Model Context Protocol (MCP) servers extend Claude's capabilities by providing access to external
 tools, APIs, and data sources. They enable Claude to interact with your development environment,
 databases, and other services while maintaining security boundaries.
 
+### Bitwarden AI plugin marketplace
+
+Bitwarden maintains a curated [marketplace of AI plugins](https://github.com/bitwarden/ai-plugins)
+specifically designed for our development workflows. This marketplace was created to provide
+quality-controlled, security-reviewed plugins that follow Bitwarden's coding standards and security
+requirements. All marketplace plugins are maintained by the Bitwarden team and include comprehensive
+documentation, testing, and security validation.
+
+To use the marketplace with Claude Code:
+
+```bash
+/plugin marketplace add bitwarden/ai-plugins
+```
+
 ### Understanding MCP servers
 
 MCP servers are separate processes that communicate with Claude through a standardized protocol.

@@ -31,6 +31,20 @@ You can find more tips for PR review here:
 
 :::
 
+## Before submitting for review
+
+Before marking a PR as ready for review, ensure you've completed the following reminders:
+
+- Contributor guidelines followed
+- All formatters and local linters executed and passed
+- Written new unit and / or integration tests where applicable
+- Protected functional changes with optionality (feature flags)
+- Used internationalization (i18n) for all UI strings
+- CI builds passed
+- Communicated to DevOps any deployment requirements
+- Updated any necessary documentation (Confluence, contributing docs) or informed the documentation
+  team
+
 ## Responding to review requests
 
 To ensure that teams within the organization operate on same set of assumptions for performing
@@ -87,6 +101,23 @@ requirements, which may not necessarily be in line with the previous solution.
 
 :::
 
+### Providing rich feedback
+
+When leaving comments on a PR, you can use emoji reactions to clarify the intent and severity of
+your feedback. This helps the author quickly understand which comments are blocking concerns versus
+informational notes or suggestions:
+
+| Emoji                                    | Meaning                                        |
+| ---------------------------------------- | ---------------------------------------------- |
+| 👍 `:+1:`                                | Excellent contribution or work                 |
+| 📝 `:memo:` / ℹ️ `:information_source:`  | Note or informational comment                  |
+| ❓ `:question:`                          | Question or inquiry                            |
+| 🤔 `:thinking:` / 💭 `:thought_balloon:` | Open for discussion                            |
+| 🎨 `:art:`                               | Suggestion or improvement                      |
+| ❌ `:x:` / ⚠️ `:warning:`                | Significant concern that should be addressed   |
+| 🌱 `:seedling:` / ♻️ `:recycle:`         | Future work or technical debt to address later |
+| ⛏ `:pick:`                              | Minor nitpick                                  |
+
 ### Review statuses
 
 When completing a review, you can either add comments, request changes, or approve the PR. It is

@@ -118,26 +118,17 @@ Initial content placeholders provide AI assistance and automated code reviews:
 
 ```
 .claude/
-├── CLAUDE.md             # General project context, guidelines, and instructions
-├── commands/             # Custom slash commands
-└── prompts/
-    └── review-code.md    # Code review prompt
+└── CLAUDE.md             # General project context, guidelines, and instructions
 ```
 
 ### Automated reviews
 
 Actions workflows for review of issues and pull requests. Targets two use cases:
 
-- Review of non-draft pull requests: The above `review-code.md` Markdown file is used as a gate to
-  execute the workflow. Repositories without this file will not have automated code reviews
-  performed. Reviews are posted via a persistent comment that updates with new commits.
+- Review of pull requests: Reviews are posted via a persistent comment that updates with new
+  commits. Reviews will only performed if an `ai-review` label is present (or added) on the pull
+  request.
 - Response: By mentioning `@claude` in issue or pull request (including inline) comments. Specific
   responses to the mention will be generated.
 
 Each above use case calls a reusable workflow in the `gh-actions` repository.
-
-### Best practices
-
-- Commands (`.claude/commands/`): For interactive Claude Code sessions.
-- Prompts (`.claude/prompts/`): For automated GitHub Actions workflows.
-- `CLAUDE.md`: General project context available in all Claude interactions.