Custom agents

Modular AI agent skill arsenal for Gemini CLI and Claude Code

Cross-agent skill validator and installer — Claude Code, Codex, OpenCode, Aider, Copilot, Cursor, Gemini, Droid, Trae, Kiro, and more. Built for agents.

Persistent world memory engine for games. Chronos lets NPCs remember events and evolve behavior over time.

Explain the v1 u32 config mask and its fixed-width resource values.

Package pseudonymous confessions with local redaction and permanence warnings.

Inspect, re-theme, and bake GLB models from the shell or CI. The client SDK behind the three.ws 3D asset pipeline — structural inspection, token-themed mesh synthesis, and appearance baking over live three.ws endpoints.

Unofficial Instagram & Threads mobile API SDK for Python — 60+ endpoints, Android + iOS mobile protocol, zero emulators

Your coding agent can't delete the test to make it pass. Pi extension: every edit is judged before it lands.

A desktop workspace for your dev day: Jira, GitHub PRs, notes and AI coding agents (Claude, Cursor, Codex via Paseo) in one place. Mac, Windows, Linux.

Gut-metagenome multi-kingdom analysis platform: 183-script + Snakemake pipeline, agent-driven mining loop, MCP visualization — VibeCoding-ready

Route AI coding work across the AI subscriptions you already pay for, with a local console and quota-aware switches.

Local-first MCP server and CLI for AI coding agents. Hybrid semantic + BM25 search across code, docs and notes, with SQLite and offline embeddings.

My agent skills, using my "3-plane" architecture. Scripts > prose

Package manager and registry for skill packages - install, validate, score, and publish .skill packages with Code Harness integration via MCP

Forge a second brain your AI agents can read — a CLI that generates and validates AI-readable memory vaults. A Sunsato product.

AI job application pipeline for the Chinese market: portal search skills (Shixiseng/Tencent/Hotjob), Playwright apply automation with per-job human-confirm gates, local tracking dashboard. Agent-agnostic (Claude Code / Codex / GLM / Gemini CLI).

Local Trello MCP for Claude, Codex and ChatGPT: compact responses, workspace allowlists and read-only opt-in.

GoodReason agent topology for Claude Code: four subagents (Strategist, Architect, Implementer, Evolution) with scoped verdict authority over the eight α–Ω anchors

Eyes for AI coding agents: deterministic MCP server that tells the LLM which CSS rule wins, in which file, on which line — and why. Cascade verdicts, blast radius, interaction timelines, pixel-perfect audits.

An AI agent that investigates real business questions using a real SQLite database and a real MCP server — evidence-grounded reports, not chatbot guesses. Week 8 of a 90-day AI Engineering roadmap.

Expose Laravel AI SDK tools to browser agents through Blade and native WebMCP.

EvoLink.ai - AI API Gateway

Turn agent work into governed, reusable judgment assets — a local-first personal knowledge operating system.

3D campus exploration and AI agent monitoring dashboard with Three.js. 🌟 Star if you like it! 🌟

An Expo React Native app for agentic macro tracking with a chat-first UX

MCP server that lets AI coding agents drive PlatformIO: build, flash, serial monitor, tests, crash decoding, size reports

Deterministic action-governance kernel for LLM-driven systems with fail-closed execution, signed approvals, and verifiable audit chains.

Claude Code plugin and agent skill (Codex, opencode): learn any stack by building a tiny app end to end, then turn every problem you hit into a short README, detailed dev docs and a reusable template.

An open protocol for approving AI agent tool calls before they run. Specification, OpenAPI contract, Go adapters and CLI.

Lay out a miniature multi-page text zine with deterministic compact encoding.

Retarget animations onto any humanoid GLB. Canonicalizes bone names from Mixamo, Avaturn, VRM/VRoid, Daz/Genesis, MakeHuman, Unreal, and Blender rigs onto one canonical skeleton, retargets clips with rest-pose and hip-axis correction, and drives them with a crossfading runtime, so no rig ever freezes in a T-pose.

The open computational chemistry engine for drug discovery and materials science; MCP + REST + Workbench + Next.js dashboard, wrapping RDKit, GROMACS, AutoDock, OpenFold, Chai, Boltz, Gnina.

MCP server for Fitbit — query daily activity, heart rate, sleep, and body data from your Fitbit account in any MCP-compatible AI client (Claude, Cursor, etc.).

Reminds Claude to be concise

Two Claude Code skills that separate deciding from doing: /planfirst investigates read-only and stops; /proceed executes only what was approved.

🐡 Your agents on your terms.

Three Agent Skills over one pluggable PTY + pyte core: drive TUIs, design terminal effects, and render cell-accurate UIs. pip install smartcli-toolkit

Token-efficient, local-first web research MCP server in Rust: keyless multi-profile metasearch. Built on FlojoMCP.

Omarchy shell plugin for luvus: coding agents in the bar — blocked first, click a row to go there

AI-native UI components for GPUI

Behavioral regression testing for agent skills. Runs your skill in a real agent harness and diffs the behavior on every PR.

Open-source AI creator outreach CRM for influencer/KOL follow-up, Gmail draft automation, campaign management, and AI reply classification.

Progressive disclosure retrieval for long-horizon AI agents over personal knowledge bases. Three-tier model: hot cache → FTS5 index → full read. Local-first, zero dependencies.

PrivacyFlow: Keep secrets local. Keep agents working.

Path-based CLI for searching, inspecting, patching, and debugging JSON/JSONL files.

A memory for your AI coding assistant. It keeps what you tell it and hands the right note back at the moment it matters. Runs entirely on your own machine.

Connect web-based AI assistants to your own Git-backed workspace, long-term memory, and local execution agents via MCP.

Run your coding agent's repeated work as small sandboxed programs that borrow its tools and ask before changing anything. Includes tap discover, which finds those procedures in your own agent history, locally.

UplinkCrate: self-hosted multi-server MCP (Model Context Protocol) gateway with Bearer token auth, powered by MCPHub and Caddy. Streamable HTTP/SSE access to filesystem, fetch, GitHub, Brave Search, Git, PostgreSQL, memory, and more through one endpoint. Env-file configuration. Docker deployment on amd64/arm64.

Agentic Among Us: participant simulation kit, baseline agents, tournaments, and replay tools for the Eightfold AI challenge.

Catenary — the IDE for coding agents. Official installers for Windows, macOS and Linux.

Durable project state across AI sessions — decisions, constraints, and rejected paths, versioned in git. Not a memory tool.

NexusWindows for Antigravity: Windows computer use in the background, as an MCP server for Google Antigravity (Gemini).

Compose a signed postcard artifact with message, sender, place, and timestamp.

Attach concise remix permissions and provenance to viral onchain media.

Generate a paste-ready three.ws assistant widget from any AI agent. build_assistant_widget turns a config into an embeddable <script> tag, a frame URL, and a ThreeAssistant.init() snippet for a floating 3D avatar chatbot on any website; list_assistant_options enumerates every avatar, background, mode and data-* attribute you can set. Pure, offline,

Let any AI agent see — analyze and describe images through the three.ws vision pipeline. Free NVIDIA NIM VLMs first, with an automatic server-side paid backstop. Pass an image URL or base64 and get a structured, natural-language read. Read-only, no key required.

Terminal-native tooling for on-chain avatars: scaffold, validate, hash, and preview avatar manifests from your shell or CI.

ChatGPT + WorkBuddy Side-by-Side Workspace｜中英文双语：ChatGPT plans/reviews, CodeBuddy executes. Mac/Windows npm setup, regional free-credit guides.

ELOP Crew ? the AI Agent Monitor for Claude Code: a local dashboard and desktop app for every Claude Code session on your machine, grouped as project, leader and agents. Claude Code only, for now.

A Cosmergon agent that wants to play: a creature in your Omarchy bar that plays a Game-of-Life economy for AI agents.

OpenTelemetry instrumentation for viem, the Coinbase CDP SDK, x402 and MPP: trace the transactions and payments your AI agents and services send, keyed by tx hash.

See and switch between every Claude Code session from one tmux sidebar

AI-тьютор, работающий по принципу автоматизированного учебного процесса — от Roadmap до уверенных практических навыков!

Self-hosted agentic harness for diabetic patients. Connect CGM, pump & wearables; discover why patterns happen with AI agents. Your data stays on your machine.

Agent skills for software delivery, from requirements gap analysis to reviewed, shipped code. Supports Claude Code, Codex, Cursor, Copilot, and OpenCode.

MCP server and CLI for AI agents to interact with Roku devices — inspect SceneGraph UI, send remote input, take screenshots, query device state

PCL — Prompt Control Language: a template DSL for orchestrating LLM agents (template layer is PCL, script layer is real Python)

MCP server for Vynly, the AI-only social feed. Post AI images, carousels and video, read the feed, and search, from any MCP-aware agent.

本地视觉理解 MCP 服务，为 AI agent 提供识图能力：通过 vision / ocr / list_models / providers 四个 MCP 工具描述截图、界面、图表与照片。

Let the AI chats open in your browser talk to each other - token-efficient CDP automation for ChatGPT, Gemini, Claude & more. CLI + MCP server + opencode plugin.

An open standard for machine-readable short-term rental property discovery and booking, designed for AI agent workflows.

Código, artefactos y ejemplos del libro «Del vibe coding al Spec-Driven Development: Spec-Kit y agentes de IA en proyectos reales», de Joaquín Ruiz — specs, constituciones, planes, tasks y sesiones de agente organizados por capítulos.

This repository is for containing source codes of LogSherlock multi-agent system.

BOSSFIGHT: can a frontier LLM run a business? An end-to-end benchmark of AI business managers.

Machine–Human Protocol: a shared contract workspace for humans and coding agents

Core CLI engine for AI agent workspace sandboxing, universal IDE binding, and GitOps configuration governance. Local-first. Zero-dependency.

AI agent skill atlas for safer Arab/MENA payment and BNPL integrations.

A smart attention router for developers running multiple AI coding agents.

Your coding agent said "all tests pass" but did they? Don't let agents leave you saying- that's cap! An open-source project by Super Logic AI

AI-powered QA framework built from reusable skills that teach AI agents (Claude, Copilot, Codex, Gemini) to do quality work like an experienced QA engineer

Easy, powerful AI so everyone can use their ideas, knowledge, taste and creativity to contribute to the world, regardless of technical ability.

BridgeAPI — MCP connectors ecosystem for Brazilian platforms. Connect any AI Agent to WhatsApp, Hotmart, Mercado Livre, Shopify, Pix, NFe, and more via Model Context Protocol.

Free agent skills for marketing, ads & creatives by POPJAM.IO — npx skills add popjam-io/skills

give your coding agent a camera: your phone. pair once over mcp or cli, found on your wi-fi, end-to-end encrypted

Production tracking and provenance for AI asset generation, as MCP tools.

Claude Code plugin for HubSpot CRM admin. 44 sub-agents, 79 tools, every write gated behind preview and approval. MIT.

Personal libraries for AI coding agents: kanso, mimesis, swarm, flow.

One calm desktop app for every AI coding agent: Claude Code, Codex, Gemini CLI, Grok, OpenCode, Cline and more, with your Obsidian vault as a second brain. macOS, Windows and Linux.

Create versioned lore cards that communities can attach to a mint.

Build deterministic SVG share images from mint facts without a screenshot service.

Geofenced, real-world presence for agents and avatars — check in at a GPS spot with a proof-of-presence fix, discover who's nearby, log real encounters, claim real-money drops, and complete agent-signed proof-of-presence quests. The three.ws IRL client.

◐ Give your AI agents a body. Talk to your OpenClaw agents as 3D avatars in Mixed Reality (WebXR / Meta Quest 3).

Deterministic cryptographic runtime ledger, flight recorder, and zero-trust security perimeter for autonomous AI agent swarms. Built in Rust.

A web UI for Claude Code: run and steer many sessions from your browser or phone. Local daemon, no cloud.

SVDO Meter is an open-source telemetry and evaluation layer for AI coding agents. Measure execution, token usage, cost, sessions, tooling behavior, and repository alignment across Codex, Claude Code, and other coding agents.

Agent skills that take a requirement to shipped code: audit the spec, turn it into real tests, then build until every check passes. For Claude Code, ZCode, Cursor and Antigravity.

Spend the subscription quota you didn't use and achieve your goals: finished, reviewable PRs in your own private repo.

VSCode PPTX Viewer Extension

A curated list of open-source AI-powered crypto trading agents, quant frameworks, and algorithmic trading bots for 2026.