- Create run-claude-core.ts as the central Claude execution function
- Update action.yml to use single "Run Claude" step
- Create src/entrypoints/run.ts that combines prepare + run-claude logic
- Pass environment variables as object to Claude instead of setting process.env
- Update create-prompt to return tools (still exports env vars for compatibility)
- Add TODO comments for future refactoring to remove env var dependency
- Both action.yml and base-action continue to work correctly
This simplifies the action flow by removing intermediate steps and environment
variable passing between prepare and run-claude phases.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Allows modes to provide custom system prompts that are appended to Claude's base system prompt. This enables mode-specific instructions without modifying the core action logic.
- Add optional getSystemPrompt method to Mode interface
- Implement method in all existing modes (tag, agent, review)
- Update prepare.ts to call getSystemPrompt and export as env var
- Wire up APPEND_SYSTEM_PROMPT in action.yml to pass to base-action
All modes currently return undefined (no additional prompts), but the infrastructure is now in place for future modes to provide custom instructions.
* feat: add slash command shipping infrastructure
- Created /slash-commands/ directory to store bundled slash commands
- Added code-review.md slash command for automated PR reviews
- Modified setup-claude-code-settings.ts to copy slash commands to ~/.claude/
- Added test coverage for slash command installation
- Commands are automatically installed when the GitHub Action runs
* fix: simplify slash command implementation to match codebase patterns
- Reverted to using Bun's $ shell syntax consistently with the rest of the codebase
- Simplified slash command copying to basic shell commands
- Removed unnecessary fs/promises complexity
- Maintained all functionality and test coverage
- More appropriate for GitHub Action context where inputs are trusted
* remove test slash command
* fix: rename slash_commands_dir to experimental_slash_commands_dir
- Added 'experimental' prefix as suggested by Ashwin
- Updated all references in action.yml and base-action
- Restored accidentally removed code-review.md file
---------
Co-authored-by: km-anthropic <km-anthropic@users.noreply.github.com>
* feat: add review mode for PR code reviews
- Add 'review' as a new execution mode in action.yml
- Use default GitHub Action token (ACTIONS_TOKEN) for review mode
- Create review mode implementation with GitHub MCP tools included by default
- Move review-specific prompt to review mode's generatePrompt method
- Add comprehensive review workflow instructions for inline comments
- Fix type safety with proper mode validation
- Keep agent mode's simple inline prompt handling
* docs: add review mode example workflow
* update sample workflow
* fix: update review mode example to use @beta tag
* fix: enable automatic triggering for review mode on PR events
* fix: export allowed tools environment variables in review mode
The GitHub MCP tools were not being properly allowed because review mode
wasn't exporting the ALLOWED_TOOLS environment variable like agent mode does.
This caused all GitHub MCP tool calls to be blocked with permission errors.
* feat: add review mode workflow for testing
* fix: use INPUT_ prefix for allowed/disallowed tools environment variables
The base action expects INPUT_ALLOWED_TOOLS and INPUT_DISALLOWED_TOOLS
(following GitHub Actions input naming convention) but we were exporting
them without the INPUT_ prefix. This was causing the tools to not be
properly allowed in the base action.
* fix: add explicit review tool names and additional workflow permissions
- Add explicit tool names in case wildcards aren't working properly
- Add statuses and checks write permissions to workflow
- Include both github and github_comment MCP server tools
* refactor: consolidate review workflows and use review mode
- Update claude-review.yml to use review mode instead of direct_prompt
- Use km-anthropic fork action
- Remove duplicate claude-review-mode.yml workflow
- Add synchronize event to review PR updates
- Update permissions for review mode (remove id-token, add pull-requests/issues write)
* feat: enhance review mode to provide detailed tracking comment summary
- Update review mode prompt to explicitly request detailed summaries
- Include issue counts, key findings, and recommendations in tracking comment
- Ensure users can see complete review overview without checking each inline comment
* Revert "refactor: consolidate review workflows and use review mode"
This reverts commit 54ca948599.
* fix: address PR review feedback for review mode
- Make generatePrompt required in Mode interface
- Implement generatePrompt in all modes (tag, agent, review)
- Remove unnecessary git/branch operations from review mode
- Restrict review mode triggers to specific PR actions
- Fix type safety issues by removing any types
- Update tests to support new Mode interface
* test: update mode registry tests to include review mode
* chore: run prettier formatting
* fix: make mode parameter required in generatePrompt function
Remove optional mode parameter since the function throws an error when mode is not provided. This makes the type signature consistent with the actual behavior.
* fix: remove last any type and update README with review mode
- Remove any type cast in review mode by using isPullRequestEvent type guard
- Add review mode documentation to README execution modes section
- Update mode parameter description in README configuration table
* mandatory bun format
* fix: improve review mode GitHub suggestion format instructions
- Add clear guidance on GitHub's suggestion block format
- Emphasize that suggestions must only replace the specific commented lines
- Add examples of correct vs incorrect suggestion formatting
- Clarify when to use multi-line comments with startLine and line parameters
- Guide on handling complex changes that require multiple modifications
This should resolve issues where suggestions aren't directly committable.
* Add missing MCP tools for experimental-review mode based on test requirements
* chore: format code
* docs: add experimental-review mode documentation with clear warnings
* docs: remove emojis from experimental-review mode documentation
* docs: clarify experimental-review mode triggers - depends on workflow configuration
* minor format update
* test: fix registry tests for experimental-review mode name change
* refactor: clean up review mode implementation based on feedback
- Remove unused parameters from generatePrompt in agent and review modes
- Keep Claude comment requirement for review mode (tracking comment)
- Add overridePrompt support to review mode
- Remove non-existent MCP tools from review mode allowed list
- Fix unused import in agent mode
These changes address all review feedback while maintaining clean code
and proper functionality.
* fix: remove redundant update_claude_comment from review mode allowed tools
The github_comment server is always included automatically, so we don't
need to explicitly list mcp__github_comment__update_claude_comment in
the allowed tools.
* feat: review mode now uses review body instead of tracking comment
- Remove tracking comment creation from review mode
- Update prompt to instruct Claude to write comprehensive review in body
- Remove comment ID requirement for review mode
- The review submission body now serves as the main review content
This makes review mode cleaner with one less comment on the PR. The
review body contains all the information that would have been in the
tracking comment.
* add back id-token: write for example
* Add PR number for context + make it mandatory to have a PR associated
* add `mcp__github__add_issue_comment` tool
* rename token
* bun format
---------
Co-authored-by: km-anthropic <km-anthropic@users.noreply.github.com>
- Switch from npm to bun for Claude Code installation in base-action
- Update Claude Code version from 1.0.59 to 1.0.61 in main action
- Ensures consistent package manager usage across both action files
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Claude <noreply@anthropic.com>
* feat: add agent mode for automation scenarios
- Add agent mode that always triggers without checking for mentions
- Implement Mode interface with support for mode-specific tool configuration
- Add getAllowedTools() and getDisallowedTools() methods to Mode interface
- Simplify tests by combining related test cases
- Update documentation and examples to include agent mode
- Fix TypeScript imports to prevent circular dependencies
Agent mode is designed for automation and workflow_dispatch scenarios
where Claude should always run without requiring trigger phrases.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Minor update to readme (from @main to @beta)
* Since workflow_dispatch isn't in the base action, update the examples accordingly
* minor formatting issue
* Update to say beta instead of main
* Fix missed tracking comment to be false
---------
Co-authored-by: km-anthropic <km-anthropic@users.noreply.github.com>
Co-authored-by: Claude <noreply@anthropic.com>
* Add mode support
* update "as any" with proper "as unknwon as ModeName" casting
* Add documentation to README and registry.ts
* Add tests for differen event types, integration flows, and error conditions
* Clean up some tests
* Minor test fix
* Minor formatting test + switch from interface to type
* correct the order of mkdir call
* always configureGitAuth as there's already a fallback to handle null users by using the bot ID
* simplify registry setup
---------
Co-authored-by: km-anthropic <km-anthropic@users.noreply.github.com>
Enables detailed permission messages in Claude Code by setting the
DETAILED_PERMISSION_MESSAGES environment variable to '1' in the
Run Claude Code step.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Claude <noreply@anthropic.com>
* feat: integrate Claude Code SDK to replace process spawning
- Add @anthropic-ai/claude-code dependency to base-action
- Replace mkfifo/cat process spawning with direct SDK usage
- Remove global Claude Code installation from action.yml files
- Maintain full compatibility with existing options
- Add comprehensive tests for SDK integration
This change makes the implementation cleaner and more reliable by
eliminating the complexity of managing child processes and named pipes.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: add debugging and bun executable for Claude Code SDK
- Add stderr handler to capture CLI errors
- Explicitly set bun as the executable for the SDK
- This should help diagnose why the CLI is exiting with code 1
* fix: extract mcpServers from parsed MCP config
The SDK expects just the servers object, not the wrapper object with mcpServers property.
* tsc
---------
Co-authored-by: Claude <noreply@anthropic.com>
Replace npm install with bun install for consistency with the rest of the project's package management.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Claude <noreply@anthropic.com>
Changed the action to cd back to the original directory after installing
dependencies, ensuring Claude runs in the context of the user's workflow
rather than the base-action subdirectory.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Claude <noreply@anthropic.com>
The action.yml was missing the fallback logic to construct the Bedrock
endpoint URL from AWS_REGION when ANTHROPIC_BEDROCK_BASE_URL is not
explicitly set. This matches the configuration in claude-code-base-action.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Claude <noreply@anthropic.com>
* feat: integrate claude-code-base-action as local subaction
- Copy claude-code-base-action into base-action/ directory
- Update action.yml to reference ./base-action instead of external repo
- Preserve complete base action structure for future refactoring
This eliminates the external dependency while maintaining modularity.
* feat: consolidate CI workflows and add version bump workflow
- Move base-action test workflows to main .github/workflows/
- Update workflow references to use ./base-action
- Add CI jobs for base-action (test, typecheck, prettier)
- Add bump-claude-code-version workflow for base-action
- Remove redundant .github directory from base-action
This consolidates all CI workflows in one place while maintaining
full test coverage for both the main action and base-action.
* tsc
* copy again
* fix tests
* fix: use absolute path for base-action reference
Replace relative path ./base-action with ${{ github.action_path }}/base-action
to ensure the action works correctly when used in other repositories.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: inline base-action execution to support usage in other repos
Replace uses: ./base-action with direct shell execution since GitHub Actions
doesn't support dynamic paths in composite actions. This ensures the action
works correctly when used in other repositories.
Changes:
- Install Claude Code globally before execution
- Run base-action's index.ts directly with bun
- Pass all required INPUT_* environment variables
- Maintain base-action for future separate publishing
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
---------
Co-authored-by: Claude <noreply@anthropic.com>
* feat: add Squid proxy network restrictions to Claude workflow
Implements URL whitelisting for GitHub Actions to prevent unauthorized network access.
Only allows connections to:
- Claude API (anthropic.com)
- GitHub services
- Package registries (npm, bun)
- Azure blob storage for caching
Uses NO_PROXY for package registries to avoid integrity check issues.
* test: add network restrictions verification test
* test: simplify network restrictions test output
* refactor: make network restrictions opt-in and move to examples
- Removed network restrictions from .github/workflows/claude.yml
- Added network restrictions to examples/claude.yml as opt-in feature
- Changed from DISABLE_NETWORK_RESTRICTIONS to ENABLE_NETWORK_RESTRICTIONS
- Added support for CUSTOM_ALLOWED_DOMAINS repository variable
- Organized whitelist by provider (Anthropic, Bedrock, Vertex AI)
- Removed package registries from whitelist (already in NO_PROXY)
Users can now enable network restrictions by setting ENABLE_NETWORK_RESTRICTIONS=true
and configure additional domains via CUSTOM_ALLOWED_DOMAINS.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Minor bun format
* test: simplify network restrictions test
- Reduce to one allowed and one blocked domain
- Remove slow google.com test
- Fix TypeScript errors with AbortController
- Match test formatting conventions
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Move network restrictions to actions.yml + show custom domains in the examples folder
* Simplify network restrictions -- Move it to actions, remove extended examples in claude.yml and move them to readme
* Remove unnecessary network restrictions test and update readme + action.yml with no default domains and respective instructions in the readme
* Update README with common domains
* Give an example of network restriction in claude.yml
* Remove unnecesssary NO_PROXY as packages are installed beforehand
* Remove proxy example -- it's intuitive for users to figure it out
* Update potential EOF not being treated as a string issue
* update claude.yml to test
* Update example allowed_domains with tested domains for network restrictions
* change to experimental allowed domains and add `.blob.core.windows.net` to use cached bun isntall
* Update remaining allowed_domains references to experimental_allowed_domains
* Reset claude.yml to match origin/main
Remove network restrictions test changes from claude.yml
* Format README.md table alignment
Run bun format to fix table column alignment
---------
Co-authored-by: km-anthropic <km-anthropic@users.noreply.github.com>
Co-authored-by: Claude <noreply@anthropic.com>
- Add settings input to action.yml that accepts JSON string or file path
- Pass settings parameter to claude-code-base-action
- Update README with comprehensive settings documentation
- Add link to official Claude Code settings documentation
- Document precedence rules for model and tool permissions
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Claude <noreply@anthropic.com>
* feat: add use_commit_signing input with default false
- Add new input 'use_commit_signing' to action.yml (defaults to false)
- Separate comment update functionality into standalone github-comment-server.ts
- Update MCP server configuration to conditionally load servers based on signing preference
- When commit signing is disabled, use specific Bash git commands (e.g., Bash(git add:*))
- When commit signing is enabled, use github-file-ops-server for atomic commits with signing
- Always include github-comment-server for comment updates regardless of signing mode
- Update prompt generation to provide appropriate instructions based on signing preference
- Add comprehensive test coverage for new functionality
This change simplifies the default setup for users who don't need commit signing,
while maintaining the option to enable it for those who require GitHub's commit
signature verification.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* feat: auto-commit uncommitted changes when commit signing is disabled
- Check for uncommitted changes after Claude finishes (non-signing mode only)
- Automatically commit and push any uncommitted work to preserve Claude's changes
- Update tests to avoid actual git operations during test runs
- Pass use_commit_signing flag to branch cleanup logic
---------
Co-authored-by: Claude <noreply@anthropic.com>
* feat: add OAuth token authentication support
Add claude_code_oauth_token as an alternative authentication method to anthropic_api_key.
This provides more flexibility for users who prefer OAuth authentication.
- Add claude_code_oauth_token input to action.yml
- Pass OAuth token through to claude-code-base-action
- Update README with OAuth token documentation and examples
- Update security best practices to cover both authentication methods
- Add OAuth example to examples/claude.yml
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* docs: add OAuth token generation instructions for Pro/Max users
Update README to mention that Pro and Max users can generate OAuth tokens
by running `claude setup-token` locally. This provides clearer guidance
for users who want to use OAuth authentication instead of API keys.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* docs: update CI capabilities documentation
- Move GitHub Actions access from limitations to capabilities in README
- Update FAQ to explain how to enable CI/CD access with actions:read permission
- Clarify that Claude can access workflow results on PRs where it's tagged
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
---------
Co-authored-by: Claude <noreply@anthropic.com>
This allows users to override the default Node version by setting the
NODE_VERSION environment variable in their workflow.
Fixes#229
Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>
Co-authored-by: Ashwin Bhat <ashwin-ant@users.noreply.github.com>
* actions server
* tmp
* Replace view_actions_results with additional_permissions input
- Changed input from boolean view_actions_results to a more flexible additional_permissions format
- Uses newline-separated colon format similar to claude_env (e.g., "actions: read")
- Maintains permission checking to warn users when their token lacks required permissions
- Updated all tests to use the new format
This allows for future extensibility while currently supporting only "actions: read" permission.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Update GitHub Actions MCP server with RUNNER_TEMP and status filtering
- Use RUNNER_TEMP environment variable for log storage directory (defaults to /tmp)
- Add status parameter to get_ci_status tool to filter workflow runs
- Supported statuses: completed, action_required, cancelled, failure, neutral, skipped, stale, success, timed_out, in_progress, queued, requested, waiting, pending
- Pass RUNNER_TEMP from install-mcp-server.ts to the MCP server environment
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Add GitHub Actions MCP tools to allowed tools when actions:read is granted
- Automatically include github_ci MCP server tools in allowed tools list when actions:read permission is granted
- Added mcp__github_ci__get_ci_status, mcp__github_ci__get_workflow_run_details, mcp__github_ci__download_job_log
- Simplified permission checking to avoid duplicate parsing logic
- Added tests for the new functionality
This ensures Claude can use the Actions tools when the server is enabled.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Refactor additional permissions parsing to parseGitHubContext
- Moved additional permissions parsing from individual functions to centralized parseGitHubContext
- Added parseAdditionalPermissions function to handle newline-separated colon format
- Removed redundant additionalPermissions parameter from prepareMcpConfig
- Updated tests to use permissions from context instead of passing as parameter
- Added comprehensive tests for parseAdditionalPermissions function
This centralizes all input parsing logic in one place for better maintainability.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Remove unnecessary hasActionsReadPermission parameter from createPrompt
- Removed hasActionsReadPermission parameter since createPrompt has access to context
- Calculate hasActionsReadPermission directly from context.inputs.additionalPermissions inside createPrompt
- Simplified prepare.ts by removing intermediate permission check
This completes the refactoring to centralize all permission handling through the context object.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* docs: Add documentation for additional_permissions feature
- Document the new additional_permissions input that replaces view_actions_results
- Add dedicated section explaining CI/CD integration with actions:read permission
- Include example workflow showing how to grant GitHub token permissions
- Update main workflow example to show optional additional_permissions usage
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* roadmap
---------
Co-authored-by: Claude <noreply@anthropic.com>
- Add fallback_model input to action.yml matching claude-code-base-action
- Pass fallback_model through to the base action
- Document the new input in README.md inputs table
- Enables automatic fallback when primary model is unavailable
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Claude <noreply@anthropic.com>
* feat: add formatted output for Claude Code execution reports
- Write turns formatter
- Modify GitHub Action to call formatter instead of dumping raw JSON
- Add comprehensive unit tests (30 tests) covering all functionality
- Add integration test with sample data for output consistency
- Support syntax highlighting for multiple content types (JSON, Python, bash, etc.)
- Include turn grouping logic and token usage tracking
- Provide CLI interface for standalone formatter usage
🤖 Generated with [Claude Code](https://claude.ai/code)
Note: seriously I have never written any line of ts code in my life, so
please make sure this is fine as I don't give any guarantees
Co-Authored-By: Claude <noreply@anthropic.com>
* Add fallback
---------
Co-authored-by: Claude <noreply@anthropic.com>
The label_trigger input was defined but not passed as an environment variable
to the prepare step, causing it to be undefined in the prepare script.
This adds the missing LABEL_TRIGGER environment variable mapping.
Co-authored-by: taku.tsunose <taku.tsunose@takutsunosenoMacBook-Pro.local>
- introduced a new input parameter `label_trigger` in `action.yml` to allow triggering actions based on specific labels applied to issues.
- Enhanced the context preparation and event handling in the code to support the new labled event.
