insane-research
insane-research, AI-powered multi-agent deep research with source verification and structured outputs. Turn a single question into a comprehensive, citation-backed research report, automatically. Quick Start • Why insane-research?
4,248 plugins found in public marketplaces and GitHub repositories, indexed and searchable. Plugin skills also install in Codex, Cursor and 4 more agents. In Claude Code, the 99 that get Auto-invokedWhat is this?This plugin ships a FLOW.md router the engine fires, so the matching skill runs itself. No slash command to remember.Learn how → fire as you prompt.
insane-research, AI-powered multi-agent deep research with source verification and structured outputs. Turn a single question into a comprehensive, citation-backed research report, automatically. Quick Start • Why insane-research?
A companion skill for the Microsoft Build conference that reads your project dependencies, finds relevant sessions, surfaces what's new for your tech stack, and helps scaffold projects and plan your schedule from the terminal.
Provides commands that preserve context across Claude Code sessions, using a session handoff to reduce context loss after compaction and mimic persistent memory in a token-efficient way.
A local desktop gateway that combines scattered LLM coding plans and API quotas into one virtual plan, connecting to Claude Code and other AI tools. Includes skills for adding providers and preparing releases.
Push it to GitHub and send us the link. We read it, then it goes live.
Sign in your way. AI works on your behalf. AI agents need access to your work systems, Jira, wikis, calendars, internal APIs. But passing credentials through shell history, environment variables, and agent context windows is a security nightmare.
Agent skills synthesizing 17 years of software engineering discipline, from Clean Code to AI-native architecture, into a single, prescriptive methodology for solo developers.
Creates black-pen illustrations, pixel-art renderings that can be animated, and deadpan watercolor illustrations with dry humor. Each style is a separate skill for image generation.
Gives an AI agent skills for Alibaba Cloud AIOps work, such as operating datasets and evaluation workflows, parsing documents, and running code in a cloud sandbox. Use it to automate cloud operations and diagnostics without writing each API call by hand.
Paper | Leandex | Demo | Putnam 2025 An agent built on Claude Code for formal theorem proving tasks. We used this system to prove all 12 problems from Putnam 2025, and completed a paper-level formalization of Effective Brascamp-Lieb inequalities.
Applies Buddhist-inspired principles as work disciplines to check your reasoning before deciding, separate what you know from what you assume, and debug with more care. Useful when weighing a choice or steadying yourself under pressure.
Skills for reviewing code diffs and PRs against a fixed checklist of correctness, edge cases, error handling and test gaps, and for writing Conventional Commits messages and splitting commit history. Also includes notes on how coding agents work internally.
Gives you a team of AI helpers for a rough product idea: one researches real data, one argues against it, and one offers cross-industry comparisons, then a coach walks the idea through to a PRD and clickable prototype.
Reverse-engineer design tokens (colors, typography, spacing, border radius, shadows) from any public website and generate starter JSON and CSS custom properties for your local project. Available as an AI agent skill (Claude, Cursor, Codex) and standalone CLI.
A multi-agent research automation plugin for Claude Code that orchestrates experiments, reviews and critiques work, and helps write scientific papers, with self-evolving skills and literature review support.
Skills, agents, and commands for Claude Code that run a one-sentence idea through a team of AI roles for research, counterarguments, and analogies, and take it to a PRD and clickable HTML prototype.
Gives an AI coding assistant a checklist to follow on each task: it clarifies what you want, picks the right capability, checks the evidence and verifies the result before it reports back. Works across Claude Code, Codex, OpenClaw and Cursor.
Walks Claude Code through the full release lifecycle: version bump, CHANGELOG update, release notes, tagging, and publishing. Triggered by requests like cut a release or publish version.
Gives Claude Code full access to Caido for web security testing. It searches HTTP history with HTTPQL, sends requests through the Caido proxy, and adds match and replace rules.
A Claude Code skill for maintaining and optimizing OpenClaw workspace files, the markdown files that form an AI agent's soul, memory, and operating procedures.
A skill for web projects that converts hex, rgb and hsl colors to OKLCH, generates palettes, checks contrast, handles gamut boundaries, and sets up themes with Tailwind.
Spec to silicon, hands off. A horse is strong, fast, and willing, and useless for heavy loads until you harness it. The harness is not a part of the horse and not a part of the cart: it is the coupling that turns raw strength into pulled weight.
Bosun is an automation plugin for autonomous engineering work, covering workflow orchestration, pull request lifecycle handling, multi-agent routing and CI quality gates. It includes skills for Playwright browser testing and for controlling native Windows desktop apps.
Automatically audits any website across three dimensions of modern search visibility: SEO, Traditional search engine optimization (Google, Bing): title tags, meta descriptions, heading structure, schema markup, internal links, content quality GEO.
AI-powered advertising strategy engine for Claude Code. Build complete ad strategies, generate platform-specific copy (Google, Meta, LinkedIn, TikTok, YouTube, Pinterest), design campaign funnels, allocate budgets, and produce professional PDF reports. 15 skills, 5 parallel agents.
Arcgentic is a harness engineering layer for AI coding agents. It turns ad-hoc prompting into a gated engineering workflow. 中文文档 -> README.zh-CN.md Arcgentic helps Codex and Claude Code run software work as a disciplined sequence: clarify the idea, plan.
CLI for the Google Health API v4, built for AI agents and developers. 40 verified data types: steps, heart rate, exercise, sleep, weight, SpO2, HRV, ECG, blood glucose, nutrition, and more Agent-first: simplified JSON output, deterministic exit codes.
MCP server that enables the use of OpenAI's high-end models and their powerful web search capabilities. By registering it with any AI coding agent, the agent can autonomously consult with OpenAI models to solve complex problems.
Gives a hard question to several AI agents that analyze it separately, challenge each other's claims over rounds, and vote toward a consensus. You get back claims that survived cross-examination and a measure of how strongly the agents agreed, useful for strategic decisions.
Gives AI coding assistants ready-made instructions for scraping data from Apify Actors, covering ads, app stores, job boards and AI search visibility. Each skill tells the assistant which Actor to run, how to build its input, and how to recover when a scrape gets blocked.
台灣第三方金流 Skills for AI Agents - 藍新金流、綠界科技 ECPay、統一金流 PAYUNi、KryptoGO 穩定幣支付 支援 Claude Code、Cursor、Codex、GitHub Copilot 等 AI coding agents。
Turns a planned change into a finished, committed code change, running planning, implementation, simplification and repeated review passes so you skip the manual loop. Each step works on the diff from your base commit to HEAD.
Turns an article into a ready-to-use cover image prompt for WeChat and Xiaohongshu covers. It reads the article, asks a few questions, suggests titles, and offers composition styles, with an option to include a real person and reference images.
Professional investment analysis frameworks for US stocks, works on every AI platform.
Any Research Input → Strategy Spec → Executable Code → Backtest → Diagnosis Turn quantitative finance research, papers, drafts, reports, or strategy ideas, into validated, executable trading strategies. Automatically.
Agent-native investment research system for evidence-tracked, refreshable investment theses. Mira is an AI investment research system for agents and research users.
A comprehensive showcase of 11 modern frontend design aesthetics with master prompts, signature effects, and production-ready code.
Helps Threads creators choose what to post next, draft it in their own voice, and estimate how it may perform based on their past results. It turns post history into a repeatable decision process, though it does not guarantee viral posts.
Gives Claude Code persistent memory across sessions. It recalls prior context, stores working state and durable facts, and logs decisions in a local SQLite database, with an MCP server and setup skill included.
A Claude Code skill that guides fine-tuning and reinforcement learning of language models with the Unsloth library, covering methods such as SFT, DPO, GRPO, ORPO, KTO and SimPO. It helps users set up their environment and train models.
Gives Claude Code persistent memory across sessions, with goal management, architecture tracking and context packets. Skills let the agent define and close goals and register software components so later sessions keep accurate context.
Reusable skills and steering that teach AI coding agents how to apply the AWS Well-Architected Framework. One set of playbooks, 15 supported tools.
Bridge Grok Build into Claude Code for review, critique, delegation, and session import. This repository is a Claude Code marketplace plugin that shells out to the real grok CLI. Run status, results, and stop are owned by the plugin (PID + log files).
A development plugin with skills for architecture review, cross-validating documentation, Mermaid diagrams, React Flow workflow visualization, and shadcn/ui components. It also supports spec-driven development, git worktrees, and pull request automation.
Turn any business idea into a complete, sellable offer, using Alex Hormozi's frameworks. A skill library for coding agents.
This skill guides you through semi-structured, interactive goal-setting using the technique of Mental Contrasting with Implementation Intentions (MCII), an evidence-based psychological exercise that draws on a self-regulation strategy to improve learning.
PYA (Protect Your Ass): security audit plugin for Claude Code. 9 specialist agents, a dated threat feed, verified findings, offline by default.
An engineering workflow layer for AI coding tools that adds specs, review steps, quality gates and traceability. Built for developers who want AI-assisted delivery to follow a structured pipeline.
A marketplace of Claude Code skills built by the Fin 2x team.
A Claude Code skill that transforms codebases into literate programs. These are documents written for human comprehension that also generate the original source code.
Gives an AI coding assistant expert guidance for low-level work such as custom memory allocators, NUMA-aware memory layout, fast packet I/O with AF_XDP and DPDK, and async I/O with io_uring. Useful when you tune systems code and want fewer trial-and-error cycles.
Browser automation plugin for Claude Code with skills for web research, form filling, file downloads, end-to-end web app testing, and accessibility audits. The LLM writes Python code that drives the browser through the Chrome DevTools Protocol.
Helps coding agents write, evaluate and optimize DSPy programs, from typed signatures and modules to metric-driven optimization and long-context reasoning. Covers the whole workflow so you can build and tune LLM pipelines without relearning the API each time.
An autonomous, looping, multi-altitude security vulnerability hunt for a codebase, packaged as a Claude Code skill.
Lets your AI agent handle Xiaohongshu (Little Red Book) for you: planning content and topics, publishing image or video notes, browsing feeds and comments, and liking, saving or replying to posts. It runs on the xiaohongshu-mcp service and works with Claude Code and OpenClaw.
Gives Claude Code skills for Sentry error monitoring: setting up the SDK, creating alerts, debugging and fixing production issues, and making stack traces readable with source maps or debug files.
A curated collection of agent skills for Claude Code and compatible agent platforms.
A library of healthcare and life sciences domain capabilities, reference agents with best practices, reusable skills, and MCP-based tools, serving builders in their IDEs, researchers in Amazon Quick, and developers deploying to AWS.
Local-first memory for AI agents and teams, with skills for shared memory across bots on one machine, KV caching of repeated reads, compressing large text to save context space, and role-based governance of team workspaces.
A browser-based platform for running CLI AI assistants such as Claude Code and Codex remotely, with session import, workspace permissions and multi-user access. Adds an office-assistant mode for planning, summaries and drafting docs or emails.
Inspect and manage UniFi network infrastructure from the terminal, covering devices, clients, VLANs, WiFi, firewall, VPN, and switch ports. Offers a command-line tool and a dashboard, usable by people and by agents.
Gives a small business a full social media team inside Claude Code, from brand voice setup to writing captions and LinkedIn posts, planning a monthly content calendar, and scheduling posts across platforms.
Groups local git branch changes into reviewable chapters viewed in a browser, and lets Claude Code pick up review comments left in the Stage UI, make the changes, and resolve each thread.
MCP-for-Stata is an safety first open-source MCP server and CLI tool with Skills lets any AI agent you use invoke local Stata on your device for regression analysis, econometrics, paper replication, and empirical research.
A universal OpenCode workflow setup with specialized agents, skills, commands, and plugins for any software project.
Note (April 2026): This project is currently on hold. I'm focusing on other work for the foreseeable future. The code works, the docs are accurate, and issues/PRs are welcome, but response times will be slower than usual. See #22 for details.
Lets a Claude Code user manage an issue board, git worktrees and parallel sessions from one place. It can bridge GitHub and Linear issues into local Claude Code runs and set up forms that email submissions.
Search WeChat public-account articles by keyword and return structured evidence without an API key. Use it for topic discovery when you know the keyword but not which WeChat public account published the useful article.
A highly opinionated plugin marketplace for augmenting your Claude Code. Battle-tested skills and tools that actually work.
A CLI tool for managing and executing SQL queries across multiple databases. Written in Go, made beautiful with BubbleTea.
Generates AI image, video and music prompts for Claude Code users, and can submit them to platforms such as Kling, Suno, Midjourney and Runway through browser automation.
Claude Code Skills for content creation, design optimization, and resource curation.
An Agent Skill and Dify plugin to transform Markdown to files of DOCX, PPTX, XLSX, PNG, PDF, HTML, MD, CSV, JSON, XML.
Scans Claude Code tool outputs such as files, web pages and command results for indirect prompt injection attempts, and warns when it finds one. Useful for developers who let Claude read untrusted content.
English · 简体中文 2dimg2motion.skill, not a filter that "makes an image move", but a runnable framework for generating game animation frames Spritesheet delivery pipeline via key-pose redrawing, identity locking, and limb topology constraints, it.
Lets the coding agents you already run in different repos, such as Claude Code, Codex and OpenCode, message each other directly, so you stop copy-pasting answers between sessions. You can also ask an agent a question, delegate a task, or check its status from a browser or phone.
A fast directory visualization tool built for AI use, with hooks and MCP tools that show a project's file structure and scan for supply chain attack risks.
Run every AI coding agent in its own git worktree, in parallel. Works with Claude Code, OpenAI Codex CLI, Gemini CLI, Aider, opencode, and any other terminal program. VibeTree is mission control for parallel AI coding agents.
Gives an AI agent long-term memory that fades unused facts over time, like human forgetting, and links related memories together so recall works across sessions. Lets your assistant remember what matters without you repeating context each time.
A library of agent skills that let Claude Code work with third-party app integrations such as 0codekit, 10Duke, 15Five and others. Each skill manages that service's data, records and workflows.
Native Rust browser automation CLI for Chrome/Chromium via CDP. gsd-browser keeps a persistent background daemon, auto-starts on first use, and exposes 92 top-level commands for navigation, interaction, authenticated live viewing, annotations, recording.
Interactive CLI tool for exploring and managing Claude Code settings and slash commands.
Helps you draft a novel inside Claude Code, with genre guidance for fantasy, mystery and romance, plus checks that catch character, world-rule and timeline contradictions as you write, and reminders when a character or plot thread has gone quiet for many chapters.
Skills for web scraping, search, and structured data extraction through Bright Data's proxies and APIs, covering agent onboarding, MCP web tools, browser session debugging, and brand listening across social platforms.
Agent skills for fal.ai. The default execution path is the genmedia CLI: discover an endpoint, inspect its schema, run it, poll status when needed, and download outputs.
Agent skills for building vector search with Qdrant, covering deployment choices, client SDKs, Edge sync, Hybrid Cloud setup, and migrating from other vector databases. Useful for developers working with coding agents.
Generates hierarchical AGENTS.md and paired CLAUDE.md files for a codebase, with complexity-scored files for subdirectories. Use it to set up agent instructions across a project.
One file. Your AI coding agent becomes a scientist. Researcher turns your coding agent into an experiment loop: you agree on a metric, and the agent tries one change at a time on its own branch, keeps what moves the number and reverts what does not.
Guardrailed video editing MCP server for AI agents. FFmpeg, Hyperframes, repurposing tools, Python client, and CLI. Local, fast, free.
Give Claude the super powers it deserves with this power user collection of bash scripts and slash commands.
Gives an AI agent live sports scores, standings, schedules, play-by-play and win probability for college basketball, college football, cricket and other leagues, plus odds math for betting analysis. Read-only data with no API keys or signup needed.
Turns a research direction into a literature survey, runnable experiments with traceable results, and a publication-ready paper, so you spend less time assembling each step by hand. Includes an integrity audit that checks papers for image misuse and numerical anomalies.
GRACE means Graph-RAG Anchored Code Engineering: a contract-first AI engineering methodology built around semantic markup, .grace XML artifacts, knowledge-graph navigation, assertions, scopes, and log-driven verification.
A Rust command-line tool that automates Ghidra reverse engineering tasks such as binary analysis, decompilation, function inspection, and binary patching. It can be used by hand or driven by an AI coding agent like Claude Code.
Skills for generating game sprite and animation frames with image generation tools, covering bustup mouth and blink frames and pixel-art character sheets for a Claude Code agent workspace. Includes memory and multi-agent features.
Gives Claude Code a team of specialized roles, such as a planner, debugger and security verifier, each with its own tools. Automatic checks catch leftover debug statements and hardcoded secrets before code reaches the main branch.
A guide to working with Claude Code in real development work, with patterns for using it from the terminal. It also bundles a skill that keeps git commit messages in a consistent conventional format.
Supercharge Claude Code and Codex with security hooks, reusable skills, a lightweight statusline, and workflow automation.
ovrtx is a C and Python library for embedding Omniverse RTX sensor simulation and visualization directly into applications.
A curated library of Seedance 2.5 video prompts plus Seedance 2.5 Skill, an installable Agent Skill that improves prompts, plans and reviews storyboards when needed, and generates controllable video.
SEOBuild Onpage - The first AI agent that writes pages Google ranks AND LLMs cite. One command in, ranking page out. Built on DeerFlow, powered by 2026 SEO + GEO strategies tested / working. Forensic competitive analysis, 500-token chunk architecture, entity consensus, verification tags.
Shows your AI coding quotas and deployment status for Claude, Codex, and Cursor, so you can check usage limits and ship status without digging through billing pages or dashboards.
Run multiple coding-agent sessions in parallel, and see which one needs you. A browser terminal for parallel AI coding agents: several sessions side by side, each in its own cell, with the one that needs you marked in colour.
Access a reMarkable library from MCP clients such as Claude, VS Code Copilot, OpenWebUI, and other compatible tools.
Harness is an AI Agent development scaffold Meta-Skill that establishes four layers of enhancement for any project in one command: knowledge management, architecture constraints, feedback loops, and entropy management.
A Claude Code skill that audits Ruby on Rails applications for code quality, test coverage and static analysis issues, following thoughtbot's Ruby Science and Testing Rails practices.
Organizes scattered AI agent skills into workflows. Lets you inspect, import, enable, disable, deploy, update, repair, and uninstall skill groups through the skill-flow CLI, across coding agents such as Claude Code and Cursor.
Claude Code skills for a solo, bootstrapped founder who builds a software business alone. The skills follow one order: sell before you build. Get the first 10 paying customers by hand, build the smallest version they need, ship it safely, then grow to 100.
Shared tooling for coding agents. Declare skills, MCP servers, hooks, subagents, and plugins in agents.toml, dotagents wires them into every agent tool on your team.
A comprehensive statusline dashboard for Claude Code. See everything at a glance, no slash commands needed.
Bring professional data and tools to the AI you already use: open-source MCP, CLI, SDKs, skills, and docs to find services, call them, and audit usage.
Skills and MCP servers for Azure development in Claude Code. They cover diagnosing live AKS incidents with read-only evidence, GPU and KAITO inference problems, bounded network captures, and building Azure portal links.
© 2026 Flowy
Built for Claude Code · Not affiliated with Anthropic
Logos are trademarks of their owners.