Gemini plugins

Orchestrate multi-LLM workflows across providers (Claude, Gemini, Codex, Copilot) for the full software lifecycle — research, architecture, implementation, code review, security audit, testing, and delivery — using structured Double Diamond phases and autonomous agents.

Implement a structured, spec-driven development lifecycle with AI agents: discuss requirements, research, plan, execute code, verify, and manage milestones. Includes multi-model plan review, automated code review, test generation, UI auditing, project documentation, and session handoff for continuous AI-assisted development.

Manage Google Ads and Meta Ads campaigns, run SEO audits, optimize content for AI search engines (GEO), and generate schema markup — all from Claude. Includes keyword research, broken link scanning, landing page scoring, and content calendar planning powered by Google Search Console and advertising APIs.

Write, optimize, and publish blog content with AI-powered research, SEO analysis, multilingual translation, and AI citation readiness. Integrates Google APIs for performance data, generates schema markup, images, audio narration, and repurposes content across platforms.

Equips Claude Code with 61 marketing and GTM skills for founders: research competitors, generate launch assets, audit pricing/pages, draft outreach sequences, analyze app store reviews, and automate content creation across blogs, social, email, and video.

Query multiple AI coding agents (Gemini, OpenAI, Grok, Perplexity, and local Claude subagents) to get diverse perspectives on architecture decisions, debugging dead-ends, and technology choices, with automated synthesis of consensus and divergence.

Enable Claude to analyze local video files or YouTube URLs with full multimodal perception: extract frames and audio via FFmpeg, detect scenes/motion/silence/transitions, transcribe speech using Whisper or Gemini, generate detailed visual descriptions, summaries, and answer questions interactively.

Run a full digital marketing agency from Claude Code — plan campaigns, execute paid/SEO/email/social, audit performance, manage brands, and generate client deliverables across 158 skills and 25 agents.

Orchestrate a multi-agent AI swarm that autonomously manages the full software development lifecycle — from brainstorming and design review through TDD implementation, PR management, code review, security audit, and release — using Claude Code, Gemini CLI, and Codex CLI with quality gates and knowledge base curation.

Turn rough dictated requests into structured prompts, critique plans with a panel of AI reviewers before committing, close sessions with handoff notes, and optionally pull second opinions from Codex or Gemini.

Run adversarial debates between multiple LLMs to cross-validate code, architecture, reviews, security, research, and planning tasks, producing specialized outputs like code, reviews, or plans.

Orchestrate cross-agent AI workflows: delegate tasks to peers, request independent implementation plans and code reviews from different AI agent backends, coordinate multi-agent communication via ask/ack and broadcast primitives, and manage repowire installation from within an agent session.

Generates business intelligence dashboards, infographics, AI images/videos, and automates posting to social media (WeChat, X, Xiaohongshu) with tools for document parsing, Obsidian integration, and browser automation.

Runs operational-misalignment diagnostics on AI agents using iFixAi, guiding you from setup through fixture authoring to inspection with configurable judges from multiple model providers.

Work with 300+ AI models through OpenRouter: query models by price/capabilities, run analytics on usage data, generate images/video/speech, transcribe audio, and integrate LLM calls into TypeScript agents with CLI tools or OAuth authentication—all without managing individual API providers.

Automate video production from storyboarding to publishing using the videoclaw CLI, with AI-powered image/video generation, TTS audio, and direct uploads to Douyin, Kuaishou, and Google Drive.

Build and orchestrate GCP data pipelines using natural language — discover assets, profile and transform data with dbt, Dataform, Spark, and BigQuery, then deploy and monitor workflows on Cloud Composer and Dataproc.

Parallel delegations to GPT, Gemini, Grok, and/or 400+ OpenRouter models for independent second opinions on code, architecture, and security — with expert subagents (architect, reviewer, debugger) and a consensus loop that iterates until all models agree. Advisory (read-only) or implementation mode.

Polish AI-generated Korean text to sound more natural and human-like. Detects 40+ AI writing patterns across multiple categories, rewrites them with surgical precision, and validates that the original meaning and facts are preserved. Supports fast and strict modes, rollback, and versioned output.

Automate local SEO workflows: manage Google Business Profiles, audit citations, monitor reviews, track map pack rankings, generate client reports, and optimize for AI search visibility. Integrates with 10+ data tools via MCP.

Run iterative development loops where Claude implements plans and OpenAI Codex independently reviews progress, providing structured feedback and gating iterations for quality assurance.

Trace, evaluate, and improve AI agents with MLflow's full agent improvement loop. Instrument Python and TypeScript code, debug individual traces and multi-turn sessions, search and query aggregated metrics, and evaluate GenAI output quality using MLflow's native APIs.

Run SEO, AEO, and GEO audits, generate structured data and content briefs, and track AI visibility across search engines and LLMs, all from within Claude Code.

Orchestrate multiple CLI-based AI agents (Claude, Cursor, Codex, Gemini) as isolated sub-agents with dedicated context, delegating specialized tasks via natural language from any host AI tool.

Polish Korean text in docs, code comments, and projects by detecting/correcting grammar, spelling, spacing, and punctuation errors; humanizing AI-generated content via 40 linguistic markers for natural style; and enforcing style consistency in tone, terminology, numbers, and formatting.

Automate end-to-end SEO workflows by connecting SE Ranking data with Planable social campaigns, generating content briefs, technical audits, backlink analysis, competitive gap reports, and AI-search optimization—all through natural language prompts

Provides 18 universal AI skills for Claude Code, enabling skill creation and optimization, deep multi-source research with citations, strategic planning using business frameworks, content processing (audio/video transcription, document conversion, web scraping), software architecture design with C4 diagrams, and orchestrated multi-step execution with review checkpoints.

Orchestrate multiple AI CLIs (Codex, Gemini, Cursor, Copilot, Qwen) as specialized subagents for planning, implementation, research, code review, and debugging, all from a single Claude Code session

Run CLI-based audits on UI design images and videos using Gemini AI to evaluate accessibility, WCAG contrast compliance, best practices, object detection, and design consistency.

Coordinate parallel queries to Claude, OpenAI Codex, and Gemini for multi-model code review and problem-solving, then synthesize their responses into a consensus verdict

Offload long-context reading, bilingual rewriting, cross-file synthesis, and second-opinion review from Claude to Google Antigravity or Gemini CLI, enabling CJK-focused document workflows and multi-AI verification.

Intelligently route AI tasks across 20+ LLM providers (OpenAI, Anthropic, Gemini, Ollama, etc.) using complexity-first model selection to minimize costs while preserving output quality. Automatically classify prompts and dispatch to the cheapest capable model, track cross-session savings, and receive weekly cost digests via Slack/Discord.

Run AI code reviews and delegate tasks to Google's Gemini from Claude Code, tracking job results and managing active background jobs.

Perform deep research on videos and web content using Gemini, then automatically generate structured reports, knowledge graphs, and explainer videos with narration and AI-generated visuals.

Route judgment-heavy SDLC work to Claude while delegating deterministic high-volume tasks like scaffolding, test generation, migrations, and first-pass code review to Antigravity (Gemini), reducing token costs and enabling hybrid agentic engineering.

Orchestrate complex engineering tasks with a cybernetics-based multi-agent pipeline: Planner specs checkpoints, Generator implements via TDD/atomic commits, Evaluator verifies with tiered checks, Retro analyzes failures for CLAUDE.md improvements. Run iterative cross-LLM code reviews with Codex/Gemini peers until consensus on fixes.

Orchestrate multi-platform marketing campaigns from Claude using the Hyper MCP — manage paid ads on Google, Meta, TikTok, Pinterest, Amazon, LinkedIn, plus organic social, SEO research, email outreach, competitor analysis, ad creative generation, and analytics integrations.

Write, analyze, and optimize blog content with anti-hallucination verification, SEO validation, multilingual translation, AI citation readiness, and quality scoring across 34 sub-skills and 8 agents.

Turn an Obsidian vault into an AI-powered second brain with daily journaling, health-metric tracking, knowledge graph generation, pattern recognition, and meeting workflows — all orchestrated through Claude Code skills and commands.

Routes AI coding tasks to the best model, enforces quality via multi-model consensus voting, and orchestrates a full development pipeline—from requirements extraction through implementation, testing, release, and documentation generation—using 46 MCP tools and a team of specialized agents.

Route tasks to the cheapest capable LLM based on complexity and type, reducing API costs across 20+ providers. Automatically classifies prompts, tracks savings, and sends weekly cost reports to Slack or Discord. An agent decomposes complex work into subtasks routed to optimal models for research, coding, and content generation.

Orchestrate Claude Code agents at scale as a technical PM: delegate coding to subagents, manage multi-phase request lifecycles, conduct parallel AI discussions and code reviews, and monitor workflows via a local dashboard—all without writing code yourself.

Automate marketing growth: run technical SEO audits across 148 rules, analyze Search Console data to fix low-CTR pages and cannibalization, build programmatic SEO pages for directories and comparisons, repurpose YouTube videos into LinkedIn posts and newsletters, create converting email sequences, brand-aligned infographics, GIFs, and PPTX case studies.

Automate full SEO marketing pipelines: scout trends/competitors on X/Twitter/Reddit, generate GEO/SEO-optimized multilingual articles with AI illustrations/voice-overs, adapt for 9 social platforms/Shopify, ingest content from YouTube/web/PDF/audio, create branded UGC short videos with lip-sync for Reels/YouTube Shorts, produce PDF lead magnets, and manage article lifecycles via Citedy API.

Run a full SEO audit of any website, covering technical SEO, content quality (E-E-A-T), schema markup, image optimization, sitemap architecture, backlink analysis, and AI search readiness, with a health score and prioritized action plan.

Run a multi-LLM discovery pipeline that chains analysis, refinement, and stress-testing across 16 AI personas to explore ideas, validate plans, and break results into actionable tasks. Includes council review, risk analysis, MVP scoping, and code generation via OpenAI, Gemini, and Claude sub-agents.

Run deep, multi-agent code reviews across LLMs (Codex, Gemini, five Claude specialist subagents) in parallel, then synthesize verified fixes for commits, PRs, or branches.

Monitor and optimize local business visibility across AI search platforms like ChatGPT, Gemini, and Google AI Overviews using geo-grid rank tracking, sentiment analysis, competitor landscape analysis, and Google Business Profile performance management.

Write complete novels with literary prose using Claude and Codex. Guide your project from story planning, character development, and drafting through line editing, critique, and EPUB ebook generation—all within the terminal.

Conduct adversarial code reviews by dispatching parallel same-model AI subagents that compete on findings via scoring, aggregating issues with worst severity on disagreements. Extend to multi-model reviews across Claude, Codex, Gemini, and others with cross-critiques to detect hallucinations and severity inflation, synthesizing deduplicated reports. Evaluate precision and recall using fixture-based suite.

Query Google NotebookLM notebooks for source-grounded, citation-backed answers powered by Gemini. Manage your notebook library—create, list, rename, delete notebooks and sources—and conduct deep research with automatic follow-ups and artifact generation (audio/video/slides) directly from Claude Code.

Generate domain-specialized images like banners, ads, YouTube thumbnails, illustrations, logos, product photos, portraits, and web assets via Google Gemini. Uses auto-built prompts, keyword triggers, format/aspect ratio detection, and reference image editing for quick design workflows.

Enables Claude Code to collaborate with multiple AI backends (Codex, Gemini, Ollama, Claude) through structured Q&A rallies, second-opinion calls, and iterative refinement rounds. Developers can delegate tasks to backend agents, get feedback loops, and synthesize results across models.

Produce AI-driven video content including product ads, short films, and e-commerce TikToks with creative direction, multimodal analysis (images/video via Gemini), direct Renoise platform CLI for media generation, and multi-platform video downloading (YouTube, TikTok, etc.)

Manage your personal and professional relationships, projects, and communications entirely locally with AI-powered insights from meeting transcripts, emails, and calendar data. Automate follow-ups, track commitments, generate dashboards, and search across all your data using natural language.

Automate BMad Method project workflows including code audits, documentation, release management, issue/PR triage, adversarial reviews, LLM prompt auditing, root cause analysis, and skill publishing to Gemini/ChatGPT.

Create Kurzgesagt-style animated explainer videos from storyboards and analyze US stock SEC filings to generate professional investment reports using AI deep research.

Perform adversarial web research and structured comparison of alternatives, with dependency-aware documentation lookups that pin versions from project manifests.

Consult Google's Gemini for second opinions on coding tasks like debugging tricky issues, architecture trade-offs, code reviews, and brainstorming alternatives. Generate images from text prompts, edit existing images, apply style transfers, compose from references, and refine iteratively using Gemini API via Python.

Route prompts to multiple LLMs (Claude, Codex, Gemini, Grok, Mistral) from a single interface, with session management, async jobs, and configurable approval strategies for parallel reviews, code review, design review, and security analysis.

Coordinate spec-driven development workflows across multiple AI models: author specs, route implementation to subagents, and run cross-model code reviews using portable Spec Kit artifacts.

Debug, test, and write cleaner code with a systematic debugging framework, test-first enforcement, testing trophy strategy, and clean code review agent. Also includes linter setup, prompt engineering, creative ideation via verbalized sampling, multi-mode analytical thinking, content analysis, and skill authoring tools.

Streamlines Claude Code development with git automation, documentation audits, command optimization, and implementation plan tracking. Includes agents for verifying completion claims and auditing code quality against project requirements.

Embed a team of 8 AI security agents in your coding workflow to automate Secure SDLC across all phases: elicit security requirements, generate STRIDE threat models, review PRs and code for OWASP/ASVS, scan IaC for misconfigs, map compliance to NIST/SOC2/GDPR, triage vulns, and gate releases with go/no-go decisions. Works in Claude Code, Cursor, Windsurf, or any MCP tool.

Orchestrate the entire development lifecycle with autonomous agents that handle task decomposition, code review, debugging, and planning while enforcing a three-red-lines discipline across 23 lifecycle skills and 20 hooks.

Train, evaluate, export, and deploy NVIDIA TAO computer vision models — covering classification, detection, segmentation, pose estimation, depth, video understanding, and 3D perception — with integrated AutoML, DEFT iterative improvement loops, synthetic data generation, and multi-platform GPU job submission (Docker, Kubernetes, SLURM, Brev, DGX Cloud).

Generate AI presentation decks from markdown or PDF reports using Gemini API. Decompose reports into structured slide JSON, then create images with 26 style presets across categories, 14 aspect ratios, and resolutions up to 4K. Supports batch deck generation and custom preset addition.

Enforce strict code-quality standards across Git workflows: distill PR review patterns into reusable rules, block merges that fail maintainability audits, fix test failures iteratively, and secure dotfiles against common security risks.

Delegate specialized AI agent tasks—code review, testing, documentation, technical research—to isolated subagents with parallel execution and context separation, while generating structured handoff documents for seamless agent-to-agent or agent-to-human transitions.

Build production-ready AI agents and multi-agent teams with the Agno framework by creating workflows and MCP integrations, adding tools, memory, debugging, and structured outputs, then deploy them via AgentOS using FastAPI.

Delegate ultra-long context (1M tokens) code analysis, deep reasoning on algorithms/architecture/optimizations, and real-time Google searches for latest framework versions/tech news to Gemini 2.5 Pro via CLI when exceeding Claude limits.

Call 143+ LLM providers through a single OpenAI-compatible interface, including chat, streaming, embeddings, search, OCR, and tool use, with a built-in proxy for key management and cost tracking.

Routes Claude Code tasks to local CLI backends (Gemini, OpenAI Codex, etc.) based on task size/type with fallback to native Claude. Includes a decompose-dispatch-verify-fix-synthesize pipeline, multi-model reasoning for consensus answers, and optional statusline HUD. Configuration via editable roster reduces costs and leverages specialized backends.

Run AI-guided design thinking sessions: provide user context via guided questions to generate a design brief, explore 5 diverse product ideas with an agent, create high-res UI mockups and prototypes using DALL-E 3 or Gemini, configure image providers, and track usage costs.

Orchestrate multi-model AI prompt execution across Claude, Codex, Gemini, and local models using isolated git worktrees and tmux sessions, with automated dependency planning, validation, and agent monitoring for complex development workflows.

Integrate Google Gemini CLI into CI/CD pipelines and automation scripts for headless code analysis, review, and validation with configurable models and JSON outputs. Orchestrate Claude-Gemini engineering loops where Claude plans and implements code while Gemini validates and reviews, streamlining complex features, security checks, performance optimization, and large-scale refactors.

Generate, edit, and composite images via Claude Code using multiple AI providers (OpenAI, Gemini, Replicate, Grok, fal, Ideogram, PhotoRoom). Configure provider and output directory, then create or modify images through natural language prompts.

Run web searches with caching, content extraction, and validation via Gemini CLI, including analytics and adversarial security checks

Spawn specialized coding subagents—coder, tester, or reviewer—powered by Claude, Codex, Gemini, or Kimi via a unified bash wrapper script. Trigger from Claude Code sessions with /spawn-coding-worker or run directly from CLI, generating session logs and markdown result artifacts for streamlined AI-assisted coding workflows.

Orchestrate Claude, OpenAI Codex, and Google Gemini in configurable feedback loops to validate mission-critical features, complex architectures, security-sensitive code, and conduct multi-AI reviews. Enable triple-AI or dual modes for thorough code quality assurance workflows.

Run multimodal image analysis, summarization, and code review by delegating to Google Gemini CLI, while also enabling Discord chat automation and server interaction through a local bridge.

Drive autonomous iteration loops for research tasks, generate publication-quality data visualizations and scientific code, fetch arXiv full-text and BibTeX citations, create images via Gemini, and stream text-to-speech.

Persist and recall AI assistant memory across sessions — capture decisions, preferences, facts, and context with typed memories, then retrieve them automatically based on current working directory or topics. Includes memory auditing for duplicates and contradictions, and encrypted passport restore for new machines.

Optimize Chrome DevTools MCP for web testing, page inspection, screenshots, and form automation, reducing token consumption 70-80% with smart text snapshots, targeted screenshots, and optional Gemini Flash vision processing.

Initialize a multi-agent AI workspace for Claude Code, Codex CLI, and Gemini CLI with shared AGENTS.md config, CLAUDE.md/GEMINI.md imports, optimized directory structure, and .gitignore separating private/public AI files—enabling seamless multi-tool AI development workflows.

Run Gemini CLI from Claude Code to review code, analyze entire codebases using its 1M token context window, and delegate tasks with the ability to retrieve detailed results including verdicts, summaries, findings, and artifacts by job ID.

Generate publication-quality academic diagrams and statistical plots from descriptions, LaTeX, or Markdown files using a Gemini-powered multi-agent pipeline. It retrieves reference examples, plans detailed specs, renders via Matplotlib or Gemini API, critiques outputs, and refines aesthetics for NeurIPS standards, enabling iterative high-fidelity research visuals.

Track costs, tokens, file changes, and git commits in AI coding sessions using Claude Code, OpenAI, Cursor, and others. Enforce budgets with warnings, check real-time status via CLI commands, and monitor analytics through a local web dashboard.

Convert docs, repos, PDFs, videos, and more into AI-ready skill packages for LLM platforms like Claude, OpenAI, and Gemini, with auto-detection of source types and configurable preset levels.

Run one-shot swarm commands using Gemini to parallel-scout codebase and external docs, synthesize XML architectural plans, and execute file implementations with quality checks and auto-fixes. Delivers fast, non-iterative coding tasks without session history or loops.

Orchestrate multi-agent Gemini pipelines for complex coding tasks: scout codebase structure and external docs for context, synthesize per-file architectural plans with user checkpoints, execute parallel implementations with iterative error fixes and code quality checks.

Submit headless jobs to LLM agents (Claude, Gemini) via Forge's MCP tools, and develop Ethereum smart contracts with Foundry's Forge CLI.

Repeat user prompts with Google Research's framing technique to boost LLM accuracy on non-reasoning tasks like multiple-choice questions, list navigation, and fact retrieval, with optional prompt filtering via PowerShell scripts.

Generate complete CNIPA-compliant Chinese patent applications from technical disclosure DOCX: orchestrate multi-agents to parse input, search prior art via Google Patents/Exa, draft abstract (<300 words)/claims/description (>10k words), create engineering diagrams (PNG), and merge into editable Word docx.

Consult Google Gemini 3 Pro for debugging, code review, architecture advice, and image generation (diagrams, infographics) directly from Claude.

Invoke Google Gemini API directly in Claude workflows to generate text and embeddings for RAG/semantic tasks, analyze multimodal content, create images via Nano Banana, execute function calls, and apply search grounding using curl or REST requests with batch support.