Add manual issue triage workflow

- Add optional issue_number parameter to Claude Issue Triage Action
- Create manual workflow that accepts comma-separated issue numbers
- Maintains backward compatibility with existing issue event workflow

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>

.devcontainer/Dockerfile

@@ -54,8 +54,8 @@ USER node
 ENV NPM_CONFIG_PREFIX=/usr/local/share/npm-global
 ENV PATH=$PATH:/usr/local/share/npm-global/bin
 
-# Set the default shell to bash rather than sh
+# Set the default shell to zsh rather than sh
-ENV SHELL /bin/zsh
+ENV SHELL=/bin/zsh
 
 # Default powerline10k theme
 RUN sh -c "$(wget -O- https://github.com/deluan/zsh-in-docker/releases/download/v1.2.0/zsh-in-docker.sh)" -- \

.devcontainer/init-firewall.sh

@@ -88,7 +88,6 @@ echo "Host network detected as: $HOST_NETWORK"
 iptables -A INPUT -s "$HOST_NETWORK" -j ACCEPT
 iptables -A OUTPUT -d "$HOST_NETWORK" -j ACCEPT
 
-# Set default policies to DROP first
 # Set default policies to DROP first
 iptables -P INPUT DROP
 iptables -P FORWARD DROP

.github/actions/claude-issue-triage-action/action.yml

@@ -12,6 +12,9 @@ inputs:
   github_token:
     description: "GitHub token with repo and issues permissions"
     required: true
+  issue_number:
+    description: "Issue number to triage (optional - defaults to event issue number)"
+    required: false
 
 runs:
   using: "composite"
@@ -32,7 +35,7 @@ runs:
 
         Issue Information:
         - REPO: ${{ github.repository }}
-        - ISSUE_NUMBER: ${{ github.event.issue.number }}
+        - ISSUE_NUMBER: ${{ inputs.issue_number || github.event.issue.number }}
 
         TASK OVERVIEW:
 
@@ -55,14 +58,22 @@ runs:
            - User impact
            - Components affected
 
-        4. Select appropriate labels from the available labels list provided above:
+        4. Identify existing duplicate issues:
+           - Use mcp__github__search_issues to find similar issues
+           - List out all open or recently closed issues that seem likely to be reports of the same problem
+           - Place the best candidate issue to merge into at the top of the list.
+             The best candidate issue is the one that is:
+               1. Has the same underlying cause as the current issue
+               2. Is the most complete and well-documented
+               3. Is the oldest open issue that is still relevant
+
+        5. Select appropriate labels from the available labels list provided above:
            - Choose labels that accurately reflect the issue's nature
            - Be specific but comprehensive
-           - Select priority labels if you can determine urgency (high-priority, med-priority, or low-priority)
+           - Consider platform labels (macos, linux, windows) if applicable
-           - Consider platform labels (android, ios) if applicable
+           - Do not apply the "duplicate" label unless you are certain the issue is a duplicate of another open issue
-           - If you find similar issues using mcp__github__search_issues, consider using a "duplicate" label if appropriate. Only do so if the issue is a duplicate of another OPEN issue.
 
-        5. Apply the selected labels:
+        6. Apply the selected labels:
            - Use mcp__github__update_issue to apply your selected labels
            - DO NOT post any comments explaining your decision
            - DO NOT communicate directly with users

.github/workflows/claude-issue-triage-manual.yml

@@ -0,0 +1,46 @@
+name: Claude Issue Triage Manual
+description: "Manually triage GitHub issues using Claude Code"
+
+on:
+  workflow_dispatch:
+    inputs:
+      issue_numbers:
+        description: 'Comma-separated list of issue numbers to triage (e.g., 123,456,789)'
+        required: true
+        type: string
+
+jobs:
+  parse-issues:
+    runs-on: ubuntu-latest
+    outputs:
+      matrix: ${{ steps.set-matrix.outputs.matrix }}
+    steps:
+      - name: Parse issue numbers
+        id: set-matrix
+        run: |
+          # Remove spaces and convert to JSON array
+          CLEANED=$(echo "${{ github.event.inputs.issue_numbers }}" | tr -d ' ')
+          IFS=',' read -ra ISSUES <<< "$CLEANED"
+          MATRIX_JSON=$(printf '%s\n' "${ISSUES[@]}" | jq -R . | jq -s -c '{issue: .}')
+          echo "matrix=$MATRIX_JSON" >> $GITHUB_OUTPUT
+
+  triage-issue:
+    needs: parse-issues
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+    permissions:
+      contents: read
+      issues: write
+    strategy:
+      matrix: ${{ fromJson(needs.parse-issues.outputs.matrix) }}
+      max-parallel: 5
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683  # v4
+
+      - name: Run Claude Issue Triage
+        uses: ./.github/actions/claude-issue-triage-action
+        with:
+          anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
+          github_token: ${{ secrets.GITHUB_TOKEN }}
+          issue_number: ${{ matrix.issue }}

.github/workflows/claude-issue-triage.yml

@@ -14,7 +14,7 @@ jobs:
       issues: write
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v4
+        uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683  # v4
 
       - name: Run Claude Issue Triage
         uses: ./.github/actions/claude-issue-triage-action

.github/workflows/claude.yml

@@ -0,0 +1,37 @@
+name: Claude Code
+
+on:
+  issue_comment:
+    types: [created]
+  pull_request_review_comment:
+    types: [created]
+  issues:
+    types: [opened, assigned]
+  pull_request_review:
+    types: [submitted]
+
+jobs:
+  claude:
+    if: |
+      (github.event_name == 'issue_comment' && contains(github.event.comment.body, '@claude')) ||
+      (github.event_name == 'pull_request_review_comment' && contains(github.event.comment.body, '@claude')) ||
+      (github.event_name == 'pull_request_review' && contains(github.event.review.body, '@claude')) ||
+      (github.event_name == 'issues' && (contains(github.event.issue.body, '@claude') || contains(github.event.issue.title, '@claude')))
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      pull-requests: read
+      issues: read
+      id-token: write
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683  # v4
+        with:
+          fetch-depth: 1
+
+      - name: Run Claude Code
+        id: claude
+        uses: anthropics/claude-code-action@beta
+        with:
+          anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
+

CHANGELOG.md

@@ -1,5 +1,142 @@
 # Changelog
 
+## 1.0.11
+
+- Claude Code can now also be used with a Claude Pro subscription
+- Added /upgrade for smoother switching to Claude Max plans
+- Improved UI for authentication from API keys and Bedrock/Vertex/external auth tokens
+- Improved shell configuration error handling
+- Improved todo list handling during compaction
+
+## 1.0.10
+
+- Added markdown table support
+- Improved streaming performance
+
+## 1.0.8
+
+- Fixed Vertex AI region fallback when using CLOUD_ML_REGION
+- Increased default otel interval from 1s -> 5s
+- Fixed edge cases where MCP_TIMEOUT and MCP_TOOL_TIMEOUT weren't being respected
+- Fixed a regression where search tools unnecessarily asked for permissions
+- Added support for triggering thinking non-English languages
+- Improved compacting UI
+
+## 1.0.7
+
+- Renamed /allowed-tools -> /permissions
+- Migrated allowedTools and ignorePatterns from .claude.json -> settings.json
+- Deprecated claude config commands in favor of editing settings.json
+- Fixed a bug where --dangerously-skip-permissions sometimes didn't work in --print mode
+- Improved error handling for /install-github-app
+- Bugfixes, UI polish, and tool reliability improvements
+
+## 1.0.6
+
+- Improved edit reliability for tab-indented files
+- Respect CLAUDE_CONFIG_DIR everywhere
+- Reduced unnecessary tool permission prompts
+- Added support for symlinks in @file typeahead
+- Bugfixes, UI polish, and tool reliability improvements
+
+## 1.0.4
+
+- Fixed a bug where MCP tool errors weren't being parsed correctly
+
+## 1.0.1
+
+- Added `DISABLE_INTERLEAVED_THINKING` to give users the option to opt out of interleaved thinking.
+- Improved model references to show provider-specific names (Sonnet 3.7 for Bedrock, Sonnet 4 for Console)
+- Updated documentation links and OAuth process descriptions
+
+## 1.0.0
+
+- Claude Code is now generally available
+- Introducing Sonnet 4 and Opus 4 models
+
+## 0.2.125
+
+- Breaking change: Bedrock ARN passed to `ANTHROPIC_MODEL` or `ANTHROPIC_SMALL_FAST_MODEL` should no longer contain an escaped slash (specify `/` instead of `%2F`)
+- Removed `DEBUG=true` in favor of `ANTHROPIC_LOG=debug`, to log all requests
+
+## 0.2.117
+
+- Breaking change: --print JSON output now returns nested message objects, for forwards-compatibility as we introduce new metadata fields
+- Introduced settings.cleanupPeriodDays
+- Introduced CLAUDE_CODE_API_KEY_HELPER_TTL_MS env var
+- Introduced --debug mode
+
+## 0.2.108
+
+- You can now send messages to Claude while it works to steer Claude in real-time
+- Introduced BASH_DEFAULT_TIMEOUT_MS and BASH_MAX_TIMEOUT_MS env vars
+- Fixed a bug where thinking was not working in -p mode
+- Fixed a regression in /cost reporting
+- Deprecated MCP wizard interface in favor of other MCP commands
+- Lots of other bugfixes and improvements
+
+## 0.2.107
+
+- CLAUDE.md files can now import other files. Add @path/to/file.md to ./CLAUDE.md to load additional files on launch
+
+## 0.2.106
+
+- MCP SSE server configs can now specify custom headers
+- Fixed a bug where MCP permission prompt didn't always show correctly
+
+## 0.2.105
+
+- Claude can now search the web
+- Moved system & account status to /status
+- Added word movement keybindings for Vim
+- Improved latency for startup, todo tool, and file edits
+
+## 0.2.102
+
+- Improved thinking triggering reliability
+- Improved @mention reliability for images and folders
+- You can now paste multiple large chunks into one prompt
+
+## 0.2.100
+
+- Fixed a crash caused by a stack overflow error
+- Made db storage optional; missing db support disables --continue and --resume
+
+## 0.2.98
+
+- Fixed an issue where auto-compact was running twice
+
+## 0.2.96
+
+- Claude Code can now also be used with a Claude Max subscription (https://claude.ai/upgrade)
+
+## 0.2.93
+
+- Resume conversations from where you left off from with "claude --continue" and "claude --resume"
+- Claude now has access to a Todo list that helps it stay on track and be more organized
+
+## 0.2.82
+
+- Added support for --disallowedTools
+- Renamed tools for consistency: LSTool -> LS, View -> Read, etc.
+
+## 0.2.75
+
+- Hit Enter to queue up additional messages while Claude is working
+- Drag in or copy/paste image files directly into the prompt
+- @-mention files to directly add them to context
+- Run one-off MCP servers with `claude --mcp-config <path-to-file>`
+- Improved performance for filename auto-complete
+
+## 0.2.74
+
+- Added support for refreshing dynamically generated API keys (via apiKeyHelper), with a 5 minute TTL
+- Task tool can now perform writes and run bash commands
+
+## 0.2.72
+
+- Updated spinner to indicate tokens loaded and tool usage
+
 ## 0.2.70
 
 - Network commands like curl are now available for Claude to use
@@ -17,7 +154,7 @@
 
 ## 0.2.66
 
-- Print mode (-p) now supports streaming output via --output-format --stream-json
+- Print mode (-p) now supports streaming output via --output-format=stream-json
 - Fixed issue where pasting could trigger memory or bash mode unexpectedly
 
 ## 0.2.63
@@ -108,4 +245,4 @@
 
 ## 0.2.21
 
-- Fuzzy matching for /commands
+- Fuzzy matching for /commands

LICENSE.md

@@ -1,3 +1 @@
-Claude Code is a Beta research preview per our [Commercial Terms of Service](https://www.anthropic.com/legal/commercial-terms). If Customer chooses to send us feedback about Claude Code, such as transcripts of Customer Claude Code usage, Anthropic may use that feedback to debug related issues or to improve Claude Code’s functionality (e.g., to reduce the risk of similar bugs occurring in the future). Anthropic will not train models using feedback from Claude Code.
-
 © Anthropic PBC. All rights reserved. Use is subject to Anthropic's [Commercial Terms of Service](https://www.anthropic.com/legal/commercial-terms).

README.md

@@ -1,53 +1,34 @@
-# Claude Code (Research Preview)
+# Claude Code
 
 ![](https://img.shields.io/badge/Node.js-18%2B-brightgreen?style=flat-square) [![npm]](https://www.npmjs.com/package/@anthropic-ai/claude-code)
 
 [npm]: https://img.shields.io/npm/v/@anthropic-ai/claude-code.svg?style=flat-square
 
-Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.
+Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows -- all through natural language commands. Use it in your terminal, IDE, or tag @claude on Github.
 
-Some of its key capabilities include:
+**Learn more in the [official documentation](https://docs.anthropic.com/en/docs/claude-code/overview)**.
 
-- Edit files and fix bugs across your codebase
+<img src="./demo.gif" />
-- Answer questions about your code's architecture and logic
-- Execute and fix tests, lint, and other commands
-- Search through git history, resolve merge conflicts, and create commits and PRs
-
-**Learn more in the [official documentation](https://docs.anthropic.com/en/docs/agents/claude-code/introduction)**.
 
 ## Get started
 
-1. If you are new to Node.js and Node Package Manager (`npm`), then it is recommended that you configure an NPM prefix for your user.
+1. Install Claude Code:
-   Instructions on how to do this can be found [here](https://docs.anthropic.com/en/docs/agents-and-tools/claude-code/overview#recommended-create-a-new-user-writable-npm-prefix).
 
-   *Important* We recommend installing this package as a non-privileged user, not as an administrative user like `root`.
+```sh
-   Installing as a non-privileged user helps maintain your system's security and stability.
+npm install -g @anthropic-ai/claude-code
+```
 
-2. Install Claude Code:
+2. Navigate to your project directory and run `claude`.
-   
-   ```sh
-   npm install -g @anthropic-ai/claude-code
-   ```
 
-3. Navigate to your project directory and run <code>claude</code>.
+## Reporting Bugs
-
-4. Complete the one-time OAuth process with your Anthropic Console account.
-
-### Research Preview
-
-We're launching Claude Code as a beta product in research preview to learn directly from developers about their experiences collaborating with AI agents. Our aim is to learn more about how developers prefer to collaborate with AI tools, which development workflows benefit most from working with the agent, and how we can make the agent experience more intuitive.
-
-This is an early version of the product experience, and it's likely to evolve as we learn more about developer preferences. Claude Code is an early look into what's possible with agentic coding, and we know there are areas to improve. We plan to enhance tool execution reliability, support for long-running commands, terminal rendering, and Claude's self-knowledge of its capabilities -- as well as many other product experiences -- over the coming weeks.
-
-### Reporting Bugs
 
 We welcome feedback during this beta period. Use the `/bug` command to report issues directly within Claude Code, or file a [GitHub issue](https://github.com/anthropics/claude-code/issues).
 
-### Data collection, usage, and retention
+## Data collection, usage, and retention
 
 When you use Claude Code, we collect feedback, which includes usage data (such as code acceptance or rejections), associated conversation data, and user feedback submitted via the `/bug` command.
 
-#### How we use your data
+### How we use your data
 
 We may use feedback to improve our products and services, but we will not train generative models using your feedback from Claude Code. Given their potentially sensitive nature, we store user feedback transcripts for only 30 days.