hermes-agent
Gives your AI agent a learning loop: it turns past work into reusable skills, keeps notes across sessions, and searches its own earlier conversations. It can also run on a cheap server and be reached from chat apps like Telegram.
Flowy lists 432 plugins for PowerShell work in Claude Code. Each one bundles skills a Claude Code agent can load while you work. Every listing shows what is inside before you install, who built it, and whether it is Auto-invoked, meaning a FLOW.md router fires the right skill as you prompt instead of leaving them dormant.
Gives your AI agent a learning loop: it turns past work into reusable skills, keeps notes across sessions, and searches its own earlier conversations. It can also run on a cheap server and be reached from chat apps like Telegram.
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by running routine tasks, explaining complex code, and handling git workflows through natural language commands.
Bundles skills for Codex users: AI image generation for visuals and sprites, a read-only code review agent, OpenAI docs lookup, and tools to create and install skills.
Turns a codebase into an interactive knowledge graph you can explore, search, and ask questions about, with skills for diff impact analysis, explanations, and business domain mapping.
A Codex plugin with skills, agents, and MCP servers for development work. Includes diagnostics, LSP code intelligence, comment checking, and tools for reporting and fixing bugs in LazyCodex and Codex CLI.
A Codex and Claude Code plugin for agent orchestration, with skills for LSP diagnostics, comment-checker feedback, health checks, and filing bug reports for LazyCodex and omo-codex.
Gives AI agents a command-line interface to real desktop and web software, so they can drive apps like personal finance, data cleanup, DNS filtering and document generation without a GUI. Generates new CLI wrappers for apps that lack one.
Control Herdr, a terminal multiplexer for coding agents, to inspect and manage panes, tabs, workspaces and commands. Herdr keeps terminals running in a background server so work survives closing the client or losing an SSH connection.
A self-hosted personal AI agent framework with skills for scheduling reminders, managing GitHub issues and pull requests, generating and editing images, searching past conversations, and installing skills from ClawHub.
An open-source personal AI assistant and agent harness that plans tasks, runs tools and skills, and keeps memory and a personal knowledge wiki. It also generates images and can create or install new skills.
A skill pack for authorized reverse engineering, security research, and penetration testing that routes tasks to the right skill and sets up needed tools on demand. It covers API security assessment and Android APK analysis, among others.
A Claude Code plugin that sets up a game development studio structure, with AI agents for different roles and workflow skills for tasks such as architecture decisions, art direction, asset audits and migration planning. It coordinates the agents in a hierarchy that mirrors a real studio.
Gives an AI agent the step-by-step workflow a senior security analyst follows, covering when to use each technique, what to check first, how to run it and how to verify the result. Covers incident response, forensics, threat intelligence, malware analysis, penetration testing and red teaming.
Gives AI agents control over Word, Excel and PowerPoint files, with skills for building .docx reports, academic papers, Excel dashboards and Morph-style animated slides.
Skills and MCP servers for driving computers with AI agents: create and clean up disposable Linux or macOS sandboxes, work inside shared Spaces and a common volume, and automate GUI tasks such as testing forms, taking screenshots, and running end-to-end QA.
Gives product managers step-by-step help with discovery, prioritization, strategy, launch and growth work, so they get structured outputs instead of loose text. Also includes skills for reviewing AI-built code before shipping.
An open-source SEO analysis plugin for Claude Code covering technical SEO, content quality, Schema.org markup, AI search optimization, local SEO and e-commerce, with specialist skills for keyword data, backlinks, image generation and site crawling.
Skills and an MCP server for building with SpacetimeDB, a relational database that runs application logic in server modules and syncs data to clients in real time. Covers the CLI and server and client SDKs for C++, C# and .NET.
A skill for installing, configuring, and troubleshooting Oh My Posh, a cross-platform prompt renderer. It covers shell setup, themes, segments, and Nerd Font icons for PowerShell, zsh, bash, and fish.
Lets Claude watch a video from a URL or local file by pulling timestamped frames and a transcript. Works in Claude Code, and the watch skill downloads with yt-dlp and extracts frames with ffmpeg.
Skills for maintainers of the googleapis/mcp-toolbox repo, covering docs link sweeps, failing test fixes, bug reproduction, pull request review, and stale issue triage. The toolbox connects AI agents to enterprise databases.
Gives coding agents a git-backed wiki for long-term project memory, with skills for retrieval, durable notes, session handoffs between agents, cross-project messaging, and memory consolidation.
IronClaw is an Agent OS focused on privacy, security and extensibility.
Gives your AI agent long-term memory that it can search across past sessions, so it recalls earlier context and preferences instead of starting from scratch. Works with DeepSeek Harness and includes a local viewer for browsing what it has stored.
Audits a website for Generative Engine Optimization, scoring how likely AI assistants like ChatGPT, Claude, Perplexity and Gemini are to cite its content. Covers AI crawler access, schema markup, brand authority and content quality, with monthly progress comparisons for clients.
An agent skill that turns complex terminal output into styled HTML pages you actually want to read. Ask your agent to explain a system architecture, review a diff, or compare requirements against a plan.
Audits paid ad accounts across platforms such as Google, Meta, Amazon, and Apple, plus budget planning and attribution checks. Built for agencies, consultants, and in-house marketing teams using Claude Code.
Gives Claude terminal control, file system search, and diff-based file editing, plus skills for setting up MCP servers, running health checks, managing knowledge bases, and organizing Obsidian vaults.
Plannotator is a local, browser-based review surface for AI coding agents: Claude Code, Codex, Copilot CLI, Gemini CLI, OpenCode, Kiro, Droid, Amp, and Pi. It plugs directly into your agent through its hooks and commands.
A skill that gives your local agent, such as Claude Code, full web access. It handles searching, page scraping, logged-in browser actions and dynamic pages through a real browser, running parallel research tasks and learning site-specific tips over time.
Configures AI coding agents such as Claude Code, Cursor, and Codex with persistent memory, skills, MCP servers, and personas. Includes skills for writing pull requests, collaboration comments, and design docs.
A Claude Code skill that decompiles Android APK/XAPK/JAR/AAR files and extracts the HTTP APIs used by the app, Retrofit endpoints, OkHttp calls, hardcoded URLs, authentication patterns, so you can document and reproduce them without the original source code.
A Model Context Protocol server that lets AI assistants control Home Assistant smart home devices in natural language, query device states, run services, and manage automations.
A local-first Agent Workbench where people and agent teams share one workspace to run multiple agent engines, manage plugins and Skills, and coordinate multi-agent projects and tasks.
This project implements a Model Context Protocol (MCP) integration between AI agent (Cursor, Claude Code) and Figma, allowing AI agent to communicate with Figma for reading designs and modifying them programmatically.
A search skill for AI agents that runs real-time web searches, vertical domain searches, parallel batch searches, and full-page content extraction. It works anonymously with lower rate limits or with an API key for higher ones.
A terminal window manager for Claude Code users, with tiling panes, workspaces, and sessions that survive restarts. The skill lets Claude read and write other panes, run work and wait on it, and report its own state.
Turns a rough feature idea into a written spec through guided questions, then has the agent build it and checks the result against that spec. Instead of hand-checking whether the work is finished, you get an automatic evaluation gate that runs in your coding agent.
Runs a math modeling contest from start to finish: it analyzes the problem, builds and solves the model, produces the code and charts, and drafts the paper in Typst or LaTeX, so you get a complete submission-ready paper instead of assembling each stage by hand.
Breaks a complex task into steps and hands each one to a specialized agent for research, coding, design or unattended automation. It also carries memory across sessions so the working setup can improve itself over time.
A self-hostable personal AI assistant that connects any model provider, builds a knowledge base from your own documents, and lets agents browse the web, run code and call MCP tools for you. Its skills cover notes, reminders, secrets and feed monitoring on macOS.
Lets an AI assistant work on your own local code, reading, editing, searching and running commands in your real projects, through a secure connection you control. Nothing is uploaded to a third party, and you approve each connection with your own password.
Plays game character voice lines and shows on-screen banners when your AI coding agent finishes or needs permission, so you can tab away without missing it. Works across many coding agents and IDEs.
Gives Hermes users a governing layer that frames each request, picks the workflow and evidence checks, and records honestly what happened. It covers planning, research, coding handoffs, operations and project memory.
A documentation plugin covering Claude Code setup, commands, workflows, agents, skills and tips, from beginner to advanced use. It also bundles skills for academic paper review, Active Directory and API security testing, and system flow tracing.
Helps you run social media posting end to end, from spotting trends and planning a content calendar to generating images, music and videos, publishing across platforms like Xiaohongshu, Douyin, Zhihu and Bilibili, and tracking what performs. You bring your own API keys for the generation tools.
Authors, validates, and troubleshoots Dagu workflow definitions in YAML and runs Dagu CLI operations. Useful for creating or debugging DAG files, choosing commands, and managing DAGs and run history.
Observal is a self-hosted registry for coding agent extensions such as MCP servers, skills, hooks and agents, with built-in usage analytics and session tracing. These skills cover registry search and publishing, operational insights, and server administration.
Gives you one workspace for running several AI coding agents side by side, with their conversations shown in a single consistent view. You can also search and resume past sessions and pass them between agents.
A Claude Code skill that enforces completion discipline on substantial autonomous work. It writes acceptance gates before execution, runs approved checks, re-verifies returned work, and reports only what the evidence supports.
Feishu/Lark channel plugin for OpenClaw, with skills for Feishu documents, cloud drive files, group chats and announcements, plus a release workflow and an end-to-end test framework for plugin developers.
Project-specific skills for the CCX AI API proxy: registering new Claude and non-Claude models, checking upstream Claude Code and Codex changes, bumping versions with CHANGELOG generation, and publishing CCX Desktop Store releases.
Lets you run DingTalk tasks from the command line or an AI agent, covering calendar and meetings, contacts, tables, documents, approvals and todos. You get structured output for scripts and agents, and an admin can preview requests before they run.
Guides Claude Code to write, fix, and refactor Go code in modern, idiomatic style, applying version-specific Go guidance from the Modern Go Guidelines CLI.
Helps Claude Code weigh evidence before judging, review its own finished work, and focus on one main task at a time when several compete. Each method comes from classic dialectical and practical philosophy.
A Claude Code skill set where Claude and Codex take turns hardening a plan through independent review, then one builds the code and the other inspects it. Use it when you want a second AI model to check work before and after it is built.
A local MCP server that lets an agent search arXiv, read individual paper sections from the original LaTeX, generate BibTeX entries, and set up topic watches. Papers are stored on disk.
A set of skills for building and running self-evolving AI agents, including ACE context-engineering loops, rich interactive reply blocks, and a company mode that runs an organization of agents with CEO, employee, and finance roles.
Lets you control Collaborator's infinite canvas from the terminal with the collab-canvas CLI, arranging tiles that show terminals, files, images, and graphs. Useful for developers building with AI agents.
A set of Claude Code skills for open-source intelligence work, covering attack-surface discovery of cloud storage buckets, continuous exposure monitoring, email domain spoofability checks, secret detection, and risk scoring of recon findings for security teams.
Manages skills, agents, hooks, plugins and MCP settings with the Skillshare CLI, syncing them across AI coding tools. Useful for teams sharing configuration and running security audits on installed resources.
Turns a local video, or an online video you have rights to process, into a subtitle quote image. It picks exact frames, crops them, and assembles a portrait or original-ratio contact sheet, keeping the original burned-in subtitles or redrawing custom ones.
Gives AI assistants and terminal users access to Korean law sources, including statutes, precedents, administrative rules, local ordinances, treaties and interpretations. It also checks legal citations for accuracy and compares statute versions over time.
Self-contained Codex and Claude Code skills for building playable, polished Three.js browser games.
Delegates coding tasks to other coding agents such as Aider, Cline, Google Antigravity, and Claude Code sessions as background implementers. The orchestrating agent then reviews each diff and lands the change itself.
Tell Claude Code what video you want, it writes the script, generates the voiceover, music, and visuals, and renders the MP4.
A Claude Code plugin with skills for Apple Notes, Apple Reminders, Bear notes, RSS and blog monitoring, and 1Password CLI secret handling. Its listed features include persistent memory and offline learning.
A collection of Claude Code skills for step-by-step problem solving, with adversarial verification of finished work, a loop that runs tasks with parallel subagents, a domain skill builder, and a release helper for configuration changes.
A Pi-native shell for controlled development with skills for opening review-sized pull requests, writing clear docs and collaboration comments, and delegating work to focused subagents with test-first habits.
Lets you manage Happier sessions from Claude Code: list, check status, send messages, wait for results, read history and stop them through the happier CLI. The sessions sync across web, desktop and mobile with end-to-end encryption.
Connects Google Search Console to your AI assistant so you can ask for SEO analysis in plain language. Covers keyword cannibalization, quick-win content targets, indexing gaps and a weekly performance report.
Gives your terminal tab completion for many CLI commands across Bash, Zsh, Fish, PowerShell, Nushell and other shells, so you can stop memorizing flags and subcommands. Skills cover the Homebrew, Bun, GitButler and Bubble Tea tools and the Carapace completion setup.
Skills for Databricks development covering the Python SDK, Databricks Connect, CLI and REST API, plus Python coding guidance on code quality, error handling, testing and environment management.
Lets Claude Code control a running Unreal Editor over MCP, covering actors, blueprints, materials, Niagara, Sequencer, widgets and more. Includes skills for building Unreal toolsets and Agent Skills.
Semantic code search and call-graph tracing for Claude Code users, letting you find code by intent, explore unfamiliar codebases, and trace callers and callees. It runs locally, and includes a skill that automates creating releases.
A skill and MCP toolkit that lets Claude Code and other AI agents drive SolidWorks CAD through natural language for parts, holes, slots, assemblies, drawings and exports. It verifies results and flags unverified capabilities for human review.
A coding agent for your terminal that navigates codebases, makes multi-file edits, runs tests and manages git. It also has skills to diagnose recurring failures, find blind spots in code and write structured critiques, and modify its own source.
Runs an autonomous agent loop that repeats a task with self-correction, task tracking and live monitoring, so a coding assistant keeps working until the job is finished. Works with Claude Code, OpenCode, Codex and GitHub Copilot.
A penetration testing automation CLI that uses AI agents to plan scans, run security tools, analyze findings and write reports. It is meant for security testers doing authorized assessments.
open-websearch provides an MCP server, CLI, and local daemon, and can also be paired with skill-guided agent workflows for live web search and content retrieval without API keys.
A skill for Claude Code that revises drafts to remove unnecessary caveats, disclaimers, hedges, and over-explanations while keeping the substance. Useful for technical and academic writing.
Gives you a desktop AI assistant that remembers your past work, keeps pursuing longer goals on its own, and can hand tasks between your devices. It can also run headless on a home server or in the cloud, and you can reach it from the chat apps you already use.
A Claude Code skill that answers questions about fund manager Zheng Xi's public investment views, quoting his own words, explains his investment method, and applies it to analyze current themes and stocks. For research and learning only, not investment advice.
A Model Context Protocol server that bridges AI models with WinDbg for crash dump analysis, user-mode remote debugging, and kernel debugging.
A library of Blender skills for AI coding tools covering animation, archviz, anime-style and horror art direction, plus an MCP server for driving Blender from Claude Code.
Gives Claude Code a shared team and org memory through Activeloop's Deeplake, with skills for tracking team goals and querying the local code graph of functions, classes, calls and imports.
Skills for KiCad electronics projects that analyze schematics, PCB layouts and Gerber files for design bugs and EMC risks, manage bills of materials, search distributors like DigiKey and Mouser, and extract specifications from component datasheets.
BaoCut: an AI video agent with an editable timeline. Transcribe, subtitle, translate, dub, cut and animate.
Lets Claude Code, Codex, Gemini CLI and other command-line AI agents send messages to each other through a local SQLite database. It needs only bash and sqlite, with no daemon or network.
A Claude Code plugin marketplace offering agent-based tools for development workflows, code analysis, code comparison, and code grading, plus skills for research, content, and residential floor plan searches.
Give Claude the ability to watch and understand videos. A Claude Code plugin that extracts frames via ffmpeg and processes audio via multiple backends (Gemini API, local Whisper, or OpenAI API).
A terminal workbench with persistent shell sessions, SSH remote work, and GPU rendering. The bundled skill lets Claude Code list workspaces and panes, send input to a pane, and capture a pane's screen.
Builds and maintains LLM-compiled knowledge bases, letting you ingest sources, compile and query a wiki, run parallel research, and track ideas and datasets. Works with Claude Code, Codex, and OpenCode.
Gives command-line tools argument completion across many shells, so Tab suggests valid flags and values for a Cobra-based CLI in Bash, Zsh, Fish, PowerShell and more. Includes skills for the completion internals of several shells.
Builds reusable AI agents and multi-agent teams for Claude Code, with skills for creating single agents, packaging existing agents, and auditing team packages with orchestrators, workers, and handoffs.
Browser automation for AI agents, turning browser actions into MCP commands or a Claude skill so agents can scrape websites, extract structured data, and control a browser with direct commands. It includes built-in scripts and a CLI, plus session cookie handling.
Runs a hierarchy of AI coding agents in parallel, with a lead agent assigning tasks to workers across tools such as Claude Code, Codex and Cursor. Includes skills for checking agent status and switching models.
GSD Pi is a local-first coding agent for planning, implementing, verifying, and tracking project work from the command line.
Searches your past AI coding sessions across Claude Code, Codex, Cursor and other agents, recalling decisions you already settled, such as naming rules, banned tools and required checks. It runs locally with no LLM calls.
Microsoft Fabric Skills are reusable AI assistant instructions for working with Microsoft Fabric. They help GitHub Copilot CLI and compatible AI coding tools understand Fabric workloads, APIs, query patterns, and operational best practices.
A spec-driven development harness for Claude Code that plans, runs, and syncs work from specifications, with quality gates, model routing, and multi-LLM cost control. Its skills cover Go project practices and a local self-evolution loop.
A Claude Code skill that simulates peer review of CS conference papers (vision, NLP, and ML venues) before submission. It can review a manuscript, give a verdict, edit the LaTeX, and check the claims.
A unified terminal tool that indexes and searches your local coding agent history from Claude Code, Codex, Gemini, Cursor, Aider and others, so past sessions can be found quickly.
Gives Claude Code a working memory of prior agent sessions. It searches the history already on your machine so you can pick up earlier work or past investigations.
Analyzes trading logs for stocks, ETFs, options, crypto and forex to judge whether returns reflect a repeatable edge or luck and survivorship bias. Includes tools to scan chats for scam wording and check guru performance claims.
A security skill suite for OpenClaw, Hermes, PicoClaw and NanoClaw agents. It provides drift detection to protect key files like SOUL.md, live security advisories, automated audits, and skill integrity checks.
Skills for building Power BI reports with Claude Code, covering Deneb and Vega-Lite visuals, custom visuals, Python and R charts, and SVG inline visuals, plus agents for semantic models, DAX and TMDL.
Installs, pins and switches between specific Zig toolchain versions using zvm (Zig Version Manager), and helps set up ZLS and resolve Zig version mismatches. Stores everything under $HOME/.zvm.
Runs SEO audits on websites, blog posts, and GitHub repositories, covering technical SEO and content checks. Built for users of Claude Code and other AI coding assistants who want repeatable, evidence-based audit results.
Luvus is a mission control tool for AI coding agents. These skills let Claude Code drive Luvus through its local CLI, delegate work to live Luvus agents or panes, and write Luvus modules that add sidebar docks and other extensions.
A design method toolkit for building and reviewing user interfaces in Claude Code. It applies consistent design tokens, layout choices, and rendered visual checks, and includes a skill for capturing approved design lessons locally.
Turns product details, packaging, target platform and audience into a ready-to-use e-commerce visual plan for Taobao, Amazon, Shopify and other marketplaces. It covers main images, detail pages, listing copy, image prompts and compliance review.
An open-source rebuild of Claude Code with skills for migrating prompts to Opus, frontend interface design, security audits, documentation updates, and root-cause debugging, plus agents, commands, and MCP servers.
A markdown-based second brain for product managers and product leads. It stores plain markdown notes in a local folder that Claude reads before answering and updates afterward, with a weekly review sweep.
Helps a research, competition or innovation project run from start to finish: finding and checking data, writing experiment code, drawing publication-ready figures, verifying citations, and keeping terms consistent across paper, slides and code.
Edits Chinese articles, reviews and documents to remove filler, repetition and formulaic phrasing, while keeping facts and the author's voice. Intended for revising existing text, not for detecting AI authorship or guaranteeing it passes AI detectors.
Finds award flights and hotel deals across many loyalty programs, comparing points prices against cash fares and checking your miles and elite status balances. Ask for a route and get a clear booking suggestion.
An open-source operating layer for Claude Code that routes requests, keeps repository state between sessions, coordinates parallel agent work, applies safety hooks, and tracks cost.
Agent skills for building Amazon, Azure and Windows machine images with Packer, pushing build metadata to HCP Packer, and writing Azure Terraform modules.