Conversation
Synced from everything-claude-code. Includes: - common/ (10 files): agents, code-review, coding-style, etc. - Language-specific rules for 12 languages: cpp, csharp, dart, golang, java, kotlin, perl, php, python, rust, swift, typescript - web/ (7 files): frontend-specific rules - zh/ (11 files): Chinese translations
Synced from everything-claude-code: code-architect, code-explorer, code-simplifier, comment-analyzer, conversation-analyzer, csharp-reviewer, dart-build-resolver, gan-evaluator, gan-generator, gan-planner, healthcare-reviewer, opensource-forker, opensource-packager, opensource-sanitizer, performance-optimizer, pr-test-analyzer, seo-specialist, silent-failure-hunter, type-design-analyzer
Converted from everything-claude-code (.md -> .toml): agent-sort, feature-dev, flutter-build/review/test, gan-build, gan-design, hookify/configure/help/list, jira, prp-commit/implement/plan/pr/prd, review-pr, santa-loop
Matching workflow files for the new commands: agent-sort, feature-dev, flutter-build/review/test, gan-build, gan-design, hookify/configure/help/list, jira, prp-commit/implement/plan/pr/prd, review-pr, santa-loop
Synced from everything-claude-code. Major additions: - Domain: healthcare-*, defi-amm-security, hipaa-compliance, customs-trade-compliance - DevOps: automation-audit-ops, dashboard-builder, github-ops, email-ops, jira-integration - AI/ML: gan-style-harness, llm-trading-agent-security, token-budget-advisor - Frontend: frontend-design, ui-demo, seo - Open Source: opensource-pipeline, openclaw-persona-forge - Video: remotion-video-creation, manim-video - And more: brand-voice, lead-intelligence, agent-sort, code-tour, etc.
This reverts commit 1a0006a.
- git mv docs/en/contributing/README.md → CONTRIBUTING.md - Update internal links to point to docs/en/contributing/ subdirectory - Update Korean contributing guide to link back to root CONTRIBUTING.md - Update root README.md to reference new CONTRIBUTING.md path
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request introduces a wide array of extensions for the Gemini CLI, including new agent definitions, slash commands, language-specific rules, and AI workflow skills. The review feedback identifies several style guide violations: agent frontmatter must remove the unsupported "model" field and use specific tool naming conventions, skill documents must use the exact header "## When to Use", and command prompts must include usage examples. Additionally, the placeholder email in the Code of Conduct needs to be replaced with a valid contact address.
- Wrap bare URL with angle brackets (MD034) - Remove trailing blank line (MD012)
… agents - Remove unsupported `model: sonnet/opus` field from all new agents - Convert Claude tool names to Gemini CLI format: Read→read_file, Write→write_file, Edit→replace_in_file, Bash→run_shell_command, Grep→search_files, Glob→list_directory
Per style guide, skill documents must use the exact header '## When to Use'.
- Replace {{ email }} with contact@jamkris.com
- Wrap email and URL with angle brackets to fix markdownlint MD034
|
/gemini review |
There was a problem hiding this comment.
Code Review
This pull request introduces a comprehensive suite of agents, commands, rules, and skills for the Gemini CLI extension, covering various programming languages and specialized domains. Feedback highlights the need to align JWT detection regexes for consistency and refactor Node.js scripts to use CommonJS instead of ESM to comply with the style guide. Additionally, branding references should be updated to 'Gemini CLI', agent references in command files must follow the '@agent-name' format, and a note in the rules documentation should be rephrased for professionalism.
Credit the original project as the basis for this Gemini CLI migration in both English and Korean READMEs.
Add {20,} minimum length to JWT header/payload segments for more
reliable secret detection.
Update conversation-analyzer and opensource-packager to reference Gemini CLI and GEMINI.md instead of Claude Code and CLAUDE.md.
…mand Per style guide, agent references use @agent-name format. Also adds example usage and related commands section.
Per style guide, Node.js scripts must use CommonJS require() not ESM import. Converts all 9 .mjs files to .js and updates SKILL.md refs.
|
/gemini review |
There was a problem hiding this comment.
Code Review
This pull request introduces several new agent definitions and skill documentation files, including the GAN-style harness and various domain-specific reviewers. My review identified several inconsistencies where references to the original project (Claude) were not fully migrated to the Gemini ecosystem, and some tool definitions in the agent files did not align with the repository's style guide. I have flagged these for correction to ensure consistency across the project.
… opensource-packager - conversation-analyzer: 'Claude' → 'the agent' (7 occurrences) - opensource-packager: 'claude' command → 'gemini' command
|
/gemini review |
There was a problem hiding this comment.
Code Review
This pull request introduces a comprehensive set of agents, commands, and skills to migrate the project to the Gemini CLI ecosystem, including new security policies and multilingual documentation. Feedback was provided regarding several style guide violations in the new JavaScript and shell script examples, specifically concerning unquoted variables, unquoted glob patterns, and functions exceeding the 50-line limit. Additionally, an improvement was suggested for a git command in the PRP commit configuration to better handle filenames with spaces.
- opensource-forker: quote $SOURCE_DIR and $TARGET_DIR - performance-optimizer: quote glob patterns for source-map-explorer - prp-commit: use null-terminated git ls-files for space-safe staging
|
/gemini review |
There was a problem hiding this comment.
Code Review
This pull request migrates the 'Everything Claude Code' ecosystem to the Gemini CLI, introducing a comprehensive suite of markdown-based agents, skills, and TOML commands. The migration includes new security policies, contributor guidelines, and specialized agents for tasks like code architecture, performance optimization, and healthcare compliance. Feedback focuses on aligning file paths with the ~/.gemini/ convention, ensuring hook scripts exit gracefully with code 0 to avoid blocking the CLI, and refactoring oversized JavaScript files to meet the 400-line limit. Additionally, suggestions were made to include the -y flag in npx commands for non-interactive environments and to improve command discoverability by adding 'Related Commands' sections to TOML prompts.
Bulk replacement of leftover Claude path references in rules/, skills/ck/, and other synced skill files.
Prevents npx from hanging in non-interactive environments.
Bulk migration of remaining Claude→Gemini references: - CLI commands: claude → gemini, claude -p → gemini -p - Paths: .claude/ → .gemini/, CLAUDE.md → GEMINI.md - Product names: Claude Code → Gemini CLI - Env vars: CLAUDE_* → GEMINI_* - Tool names: continuous-claude → continuous-gemini Preserved external references (Anthropic API model IDs, Claude Desktop, external URLs) that refer to actual third-party products.
- claude-api skill: full rewrite to Google Generative AI SDK - claude-devfleet: renamed to gemini-devfleet - Model IDs: claude-sonnet → gemini-2.5-flash, etc. - browser-qa: removed claude-in-chrome, use Playwright/Puppeteer - mcp-server-patterns: Claude Desktop → Gemini CLI - agent-eval: claude-code → gemini-cli - santa-loop: removed Claude Opus from Reviewer B, keep GPT-5.4 - cost-aware-llm-pipeline, regex-vs-llm: updated model IDs Only remaining "Claude" is in an external GitHub URL (token-budget-advisor).
Update English and Korean docs to cover all newly synced items: - agents: analysis, GAN harness, open source, healthcare, SEO - commands: feature-dev, review-pr, GAN, hookify, PRP, flutter, jira - skills: DevOps, domain-specific, architecture, media, business
|
/gemini review |
There was a problem hiding this comment.
Code Review
This pull request completes the migration from the Claude Code ecosystem to the Gemini CLI ecosystem, updating documentation, command definitions, and skill files. I have reviewed the changes and identified a few areas where the migration was incomplete or inconsistent with the project's style guide, specifically regarding environment variables, file path limits, and command examples. Please address these points to ensure full compatibility with the new environment.
- rules/dart/hooks.md: CLAUDE_FILE_PATHS → GEMINI_FILE_PATHS - autonomous-agent-harness: Anthropic dispatch URL → Google API - continuous-learning: CLAUDE_TRANSCRIPT_PATH → GEMINI_TRANSCRIPT_PATH
- opensource-pipeline: CLAUDE.md → GEMINI.md (6 occurrences) - product-lens: CLAUDE.md → GEMINI.md in workflow step - skill-stocktake: CLAUDE.md → GEMINI.md in checklist (2 occurrences) - ck/session-start.js: CLAUDE.md → GEMINI.md in comment
|
/gemini review |
There was a problem hiding this comment.
Code Review
This pull request completes the migration from the Claude Code ecosystem to the Gemini CLI ecosystem, updating documentation, command definitions, and skill configurations. I have reviewed the changes and provided feedback on minor inconsistencies in configuration file paths and command references. The changes are well-structured and align with the project's new identity.
Replace undefined PORT with ${PORT:-3000} fallback.
|
/gemini review |
There was a problem hiding this comment.
Code Review
This pull request completes the migration from the Claude Code ecosystem to the Gemini CLI ecosystem. It includes the addition of new agents, skills, and documentation, as well as updates to existing commands and configuration files to reflect the new branding. I have reviewed the changes and provided feedback on consistency and coding standards.
1 participant
No description provided.