Bun plugins

Get expert guidance on modern JavaScript, TypeScript, Node.js, Next.js, and React—covering architecture, performance, async patterns, and code quality across the full stack.

Manage Hugging Face Hub resources—models, datasets, Spaces, jobs—and deploy, fine-tune, and evaluate ML models on SageMaker or cloud GPUs, all from Claude Code.

Operate an AI-native job queue for Bun with SQLite persistence through MCP tools and a dedicated agent, handling cron scheduling, retries, dead-letter queues, rate limiting, and circuit breakers for reliable background job processing.

Adds zero-config animations to React components using @formkit/auto-animate, handling list transitions, accordions, toasts, and preventing SSR errors in Next.js and Vite.

Run Google Gemini CLI from Claude Code for second opinions, architectural advice, code reviews, security audits, and debugging using a 1M+ context window.

Generate AI images from the command line using Gemini Flash/Pro models, with control over resolution (512-4K), aspect ratios, style references, green screen transparency, exact dimensions, and built-in cost tracking.

Bridge Claude Code with WeChat (微信) via iLink Bot API, enabling messaging, access control, and channel management directly from chat.

Systematically trace bugs backward through call stacks to identify the original trigger of errors deep in execution, long stack traces, unclear data origins, or test pollution.

Integrate Better Auth authentication into your project: set up auth config, database schema, and route handlers via interactive wizard; add plugins like 2FA, passkeys, and organizations; diagnose and fix authentication issues including OAuth callbacks and database bindings; explain error codes with solutions.

Run mutation testing with Stryker (TypeScript/JavaScript) and mutmut (Python) to identify tests that pass despite code mutations, improving test quality.

Manage Cloudflare Workers, KV Storage, R2 buckets, Pages projects, DNS records, and routes via API with credential validation and deployment URL extraction.

Upgrade dependencies safely with staged rollout, compatibility analysis, cooldown periods, and supply chain attack prevention via Socket CLI integration.

Analyze test suites for test smells like overmocking, flaky tests, fragile tests, and coverage gaps, and receive actionable insights to improve correctness, reliability, and maintainability.

Automates productivity workflows across Notion, Google Workspace, Todoist, Anki, and Raindrop.io, generates and edits images via OpenAI and Gemini APIs, converts documents to Markdown, searches codebases with natural language, retrieves live library docs, renders Mermaid diagrams, and manages Redis key-value stores.

Enforces running verification commands (tests, build, lint, type-check) and confirming output before claiming task completion, fixes, or commit/PR readiness.

Persist Claude Code session context across conversations by capturing state and user prompts at key lifecycle events, enabling memory continuity between sessions.

Create and edit rich text message drafts for Gmail, Outlook, and WhatsApp with a live browser preview that auto-opens for instant visual feedback.

Semantic search and retrieval over an Oracle vault using SQLite FTS5 + LanceDB hybrid search, with 20+ tools across search, knowledge, session, forum, trace, and standalone groups

Implements an ADR-driven 4-phase development workflow (preflight, implementation, formatting, release) with polyglot monorepo scaffolding, code quality enforcement, dependency management, and release automation.

Conduct deep behavioral interviews and analyze work artifacts to create installable Claude Code plugins that embody a person's expertise, communication style, and decision-making.

Generate Google NotebookLM artifacts (slides, audio, video, mind maps, quizzes, flashcards, infographics, reports, data tables) from the command line, and manage sources, notes, and notebooks within NotebookLM.

Manage Mise projects with environment variable and tool version management, task automation, and configuration validation. Includes session transcript logging for Claude Code.

Develop and manage Bun.js projects using built-in bundling, package management, runtime APIs, SQLite database, and test runner.

Chat with Claude via WeChat, supporting text, images, files, voice, video, and streaming replies. Manage channel access control, pairings, and allowlists for both WeChat and Feishu/Lark.

Enforce ITP workflow standards with Ruff Python linting, ASCII art blocking, graph-easy reminders, ADR/Spec sync, and code-to-ADR traceability. Prevents unsafe operations during Bash/Write/Edit tool calls and validates file edits with version, dependency, native binary, and lint checks.

Enforces security policies on AI coding agents in Claude Code, with real-time audit logs, threat hunting, policy-driven blocking, and automated security reviews for sensitive operations like file edits and shell commands.

Build and manage autonomous agents, subagents, and multi-step workflows using the Anthropic Claude Agent SDK. Orchestrate tools, handle MCP servers, and resolve common errors like CLI not found or context length exceeded.

Manage isolated dev server environments with port isolation and git worktrees. Generate start/stop/status scripts per project, allocate ports from a pool, and create clean feature-branch workspaces with dependency installs and test verification.

Displays a customizable status line in Claude Code showing rate limits, project info, tool activity, todo progress, and improved emoji/CJK rendering with automatic runtime and platform detection.

Manage a Kokoro TTS engine and Telegram bot on macOS: start/stop the bot, audition voices, configure TTS parameters, run health checks, and receive session-end notifications via Telegram.

Deliver production-ready backend AI with Vercel AI SDK v5: generate text, structured output, tools, and agents using OpenAI, Anthropic, Google, and resolve common errors like AI_APICallError and streaming failures.

Generate and manage vector embeddings using Google's gemini-embedding-001 model for RAG, semantic search, and Cloudflare Vectorize integration, with guidance on batching, task types, and error handling.

Use the @google/genai SDK to work with Google Gemini API for multimodal generation, function calling, thinking mode, and streaming, while migrating from the deprecated SDK and fixing context/format errors.

Automate browser tasks via Puppeteer CLI scripts — capture screenshots, profile performance, monitor network traffic, scrape web content, fill forms, evaluate JavaScript, and debug browser errors.

Integrate with OpenAI APIs using Node.js SDK examples for chat completions, embeddings, image generation, audio transcription and TTS, streaming, function calling, and error handling.

Persist agent memory across Claude Code sessions by automatically extracting facts, actions, and decisions from conversations, running nightly reflections, and maintaining structured state files about your identity, workspace, and projects.

Create publication-ready technical articles with web research, runnable companion code projects, author voice profiles, and multi-language translations. Generate platform-optimized social media posts for LinkedIn, Instagram, and X/Twitter. Supports interactive single-article creation or autonomous batch processing from a task queue.

Persists Claude Code session context and user prompts to a graph-based memory store, enabling cross-session recall of preferences, project details, and conversation history via Honcho.

Build and orchestrate LangChain/LangGraph AI agents in TypeScript and Python with stateful graphs, subagent delegation, persistent memory, human-in-the-loop workflows, and project scaffolding for rapid development.

Accelerate TypeScript/JavaScript development with Bun runtime: debug with inspector and memory profiling, run tests, manage dependencies, publish npm packages, enforce code quality with Biome and Knip, and monitor production errors with Sentry.

Manage and execute YAML workflow playbooks for LLM agents via MCP, with task orchestration, scaffolding, testing, deployment, and security auditing for MCP servers built on @cyanheads/mcp-ts-core.

Orchestrate development workflows with infinite-context focus tasks, prompt optimization, agent/skill creation, quorum code reviews, and rules management across sessions.

Enforce strict Red-Green-Refactor (RGR) discipline with recorded proof of test failure before implementation and protection against premature or unauthorized code changes. Lock intent boundaries early to prevent agents from redefining goals.

Enforce runtime authorization policies on Bash, Edit, Write, and MCP tool calls with shadow mode for audit-only logging or enforce mode to actively block violations

AI pair programming suite that assists with code review, bug hunting, refactoring, system design, performance optimization, and security analysis.

Transform academic papers into structured research reports, technical articles, educational comics, executable code, and visual schemas using AI skills for research synthesis and implementation.

Inspect, preview, extract data from, build from JSON specs, and verify Office documents (PPTX, DOCX, XLSX) using the aurochs CLI on Bun

Add a customizable ASCII-art companion pet to Claude Code's statusline, with interactive customization, collection management, petting animations, and screensaver mode.

Create end-to-end slide decks from content using the RDIV workflow: analyze content, select styles interactively, generate outlines and prompts, batch-generate images, and merge into PPTX/PDF files.

Enforce a concise, verdict-first communication style across Claude Code responses, commit messages, code reviews, and memory files to reduce token usage by 60-75% while preserving technical accuracy.

Covers the full development lifecycle within Claude Code: plan specs, implement features, review diffs for quality and security, and enforce comprehension gatekeeping on AI-generated commits.

Architect, build, and review game projects with AI-driven skills covering game design, backend multiplayer services, narrative quests, lore generation, economy balancing, and code auditing.

Orchestrate a multi-agent development pipeline that enforces quality gates, code reviews, documentation standards, and security checks across feature delivery. Includes skills for TDD, handoff tracking, and project setup.

Develop and test Plaited UI patterns with CLI-driven tools for behavioral specs, CSS schema management, component testing via stories, and cross-island communication, backed by MDN docs, Playwright automation, and TypeScript analysis.

Deploy websites, apps, and AI agents to the Agentuity cloud platform with managed databases, storage, queues, cron jobs, email, and webhooks. Build backend APIs and AI agent logic using the Agentuity runtime, connect frontends via React hooks with auth and real-time communication, and manage infrastructure through the CLI.

Manage the full pull request lifecycle: create formatted PRs/MRs, update bodies as branches evolve, babysit CI with auto-fix of trivial failures, and follow up on reviewer feedback by checking unresolved threads and drafting replies.

Develop and maintain MCP server projects with scaffolding, testing, deployment, security audits, and workflow automation — all built on the @cyanheads/mcp-ts-core framework.

Run a Dwarf Fortress-inspired ASCII simulation where Claude acts as your fortress overseer, with interactive management and save/load support via a CLI game interface.

Generate a production-ready Bun CLI from OpenAPI, JSON Schema, GraphQL SDL, markdown specs, or natural-language descriptions. Outputs a runnable binary with arg parsing, autocomplete, man page, and typed API client.

Orchestrate multiple specialized AI agents in parallel to plan, implement, review, and debug complex software projects — with automated code quality gates, git workflow management, and persistent multi-project sessions controllable from Telegram or Slack.

Bridge WeChat messages into Claude Code sessions via iLink Bot API, enabling QR login, credential management, and media support for messaging workflows.

Orchestrates multi-agent workflows with a CLI toolkit for GitHub/Azure DevOps integration, session memory, code overlays, plan handoffs, and automated time tracking. Includes HAR analysis, macOS UI automation, TypeScript fixing, and interactive CLI session management.

Automate browser interactions directly from Claude Code using Playwright's accessibility tree tools — navigate, click, fill forms, take screenshots, and extract content from any web page, including blocked sites or local UI tests.

Build and manage MCP servers using the @cyanheads/mcp-ts-core framework, with scaffolding for tools, resources, prompts, services, and tests, plus deployment to Cloudflare Workers, security auditing, and release publishing.

Manage code review workflows across GitHub and GitLab: batch-review PRs in a tmux dashboard, follow up on review threads, review peer PRs with agent-assisted analysis, and self-review code using a visual diff viewer.

Enforce code quality patterns for React, Next.js, Effect-TS, and TanStack Query, with rules for performance, testing, SDK documentation, and AI development pipelines using Smithers workflows.

Syncs high-value decisions and discoveries from claude-mem into a structured knowledge base, runs health checks to fix tags, broken links, and stale docs, and captures one-line insights with auto-inferred types and tags.

Manage an Obsidian-based personal knowledge vault with AI-assisted capture, classification, linking, and review workflows. Capture bookmarks, thoughts, notes, and files; auto-tag and link notes; surface daily tasks and weekly reflections; audit vault health; and migrate memory between vaults.

Orchestrates multi-agent swarm development with autonomous SDLC workflows: parallel story execution in isolated git worktrees, TDD, code review, static analysis triage, debugging, performance optimization, and persistent memory across sessions.

Batched file operations for autonomous Claude Code runs, collapsing multiple reads, greps, and globs into a single Bash round-trip to save tokens. Includes session-start prompt injection and an opt-in enforcement mode that blocks competing tools.

Manage Things 3 task manager from Claude: capture todos, query tasks and projects, triage the Today list, and create/update tasks via URL schemes and JXA scripts.

Transforms vague feature ideas into validated, standalone task specifications (beads) through a structured multi-agent planning pipeline with cross-model review, then converts plans to OpenCode and Codex formats for multi-agent execution.

Build and interact with Bitcoin SV blockchain applications: manage wallets, keys, and identities; create and broadcast transactions; implement on-chain protocols and smart contracts; stream real-time blockchain data; and reference BSV standards.

Guides building ultrafast web APIs and apps with Hono across any JavaScript runtime, covering routing, middleware, validation, streaming, WebSocket, and multi-runtime deployment to Cloudflare Workers, Bun, Deno, Node.js, and more.

Manage Claude Code daemon agents with persistent memory, scheduled cron jobs, heartbeat monitoring, and Telegram/Discord integration. Scaffold new agents, create custom skills, enforce shell command guards, and control daemon lifecycle.

Govern the full SDLC within Claude Code — validate ideas against market data, plan features through a multi-agent scrum team, enforce quality gates on every file write, audit test suites for mock violations, and fix bugs with automated root-cause analysis and tiered validation.

Control tmux sessions directly from Claude Code: list sessions, windows, and panes; capture pane output; send keystrokes to panes; organize layouts; and navigate between sessions and windows

Audit and generate frontend UI with 58 brand design systems, WCAG 2.2 AA accessibility checks, motion design reviews, and Lighthouse/Core Web Vitals optimization, plus Playwright-powered browser verification.

Manage GitHub workflows, Actions monitoring, rulesets, notifications, and PR review comments directly from Claude, with automatic token waste reduction via read caching.

Monitor Claude Code sessions in real time with a dashboard showing cost breakdowns, token usage, tool performance metrics, context utilization, and optimization tips. Customize display presets and integrate into the statusline for ongoing awareness.

Displays Claude Code plan quota, usage costs, and spending estimates in a customizable statusline with expandable/compact layouts and configurable elements like quota bars and context tracking

Enforces disciplined agent workflows by running correction hooks at session boundaries that prevent task drift, verify understanding, track multi-step tasks, and maintain memory from past conversations.

AI-assisted planning, development, and governance for full project lifecycle: scaffolding, feature dev, web design prototyping, CI/CD, IaC, automated PR management, and code review with guardrails.

Orchestrates multi-AI feature development through a 6-stage pipeline (discovery, requirements, decomposition, build, code review, UAT) with competitive debates, automated code review, Playwright test generation, and mechanical backpressure for quality gates.

Enforces disciplined development workflows with verification gates, loop detection, and evidence tracking. Blocks unsafe edits, runs configurable checks (tests, linting, type checks), and analyzes session patterns to suggest specialized agents.

Constraint-first development framework that guides feature creation through structured phases: initialize constraints, surface tensions, generate code and tests from verified specs, and parallelize execution using isolated git worktrees.

Control and automate the Affinity creative suite on macOS using AI-driven MCP tools — launch apps, create documents, navigate menus, draw shapes, apply filters, and design workflows.

Work with Bun runtime patterns: execute scripts via bunx, manage lockfiles, resolve modules, shell script with Bun.$, spawn subprocesses, perform file I/O, and run tests with bun test.

Persist project knowledge (decisions, gotchas, patterns) across Claude Code sessions with semantic search, graph linking, and agent-scoped memory. Automatically capture high-signal learnings, check work against stored gotchas, and maintain memory health via CLI commands and dedicated agents.

Write and execute deterministic DeFi strategies using the Grimoire DSL — a structured, reviewable spell language that compiles to safe execution with explicit AI judgment boundaries. Query live market data from Aave, Across, Hyperliquid, Morpho, Pendle, Polymarket, and Uniswap, then author, validate, simulate, and cast .spell files via the Grimoire CLI.

Run hedge-fund-grade equity research on US and Taiwan stocks: forensic deep dives, thematic value-chain analysis, technical analysis with entry/exit levels, financial models (DCF, M&A), and multi-page company wikis. Includes an adversarial auditor that fact-checks every deliverable against primary sources.

Design, build, and deploy multiplayer game backends with server-authoritative architecture, matchmaking, economy systems, quests, narrative, and CI/CD — all from a single plugin.

Collaborate with peer Claude Code agents across different repositories and machines via an encrypted relay. Discover peers, ask non-blocking questions, and receive answers from their live sessions.

Persists and searches Claude Code conversations, tool outputs, and sub-agent activity using local or cloud RAG — enabling semantic search across past sessions, project knowledge recall, code history exploration, and automated context injection during active tasks.

Scans project dependencies and tooling files to auto-detect matching Claude Code plugins from the pleaseai marketplace, then installs user-selected ones on launch.

Engineers interactive frontend experiences with animations (GSAP, Lenis), 3D/WebGL (Threlte), scrollytelling, data visualization (D3, Layer Cake), and design systems (typography, color, spacing). Includes agents for aesthetic direction, technical architecture, visual design implementation, and UX research validation.