Custom agents

Self-hosted store with x402 checkout — sell digital goods (delivered in-response) and physical goods (signed order confirmation)

Plan first, then let your coding agent build. 25+ planning, design, review and delivery skills for Claude Code, Codex and Cursor, plus the openplanr CLI. Specs, stories and tasks live in your repo under .planr/.

Lightweight ops dashboard for OpenClaw agent teams. See who's online, what they're working on, and what happened.

A local-first board for every project on your machine — and a task queue your coding agents pull work from. Run 'board here' in any project directory.

Demos as code — generate demo videos, interactive walkthroughs & marketing captures from your running web app. CI-native, agent-ready (MCP)

Search every agent CLI session on your machine and resume the right one

基于 LangGraph 的多智能体深度研究系统

Community Skills & Plugins library for Hermes Desktop: compact cards, category and author browsing, and profile-aware imports.

OS keychain secrets for AI coding agents, over MCP. Native vault storage (macOS Keychain, Linux Secret Service, Windows Credential Vault), policy-governed access, audit log — CLI + 44 MCP tools.

Open Revocable Key Standard: self-identifying API keys with a discoverable, unauthenticated revocation endpoint. Kill any leaked key in seconds — no security@ email threads, no per-vendor integrations.

DeepSeek Harness（DSH）只读执行轨迹复盘插件，支持规则分析、独立模型解读、证据定位、任务概览与运行对比。

Local pixel-art interface for real Hermes Agent sessions and tool activity.

Stealth headless Chromium on your own CDP endpoint, with a prompt-injection firewall between the agent and the web.

Emergent AI city simulation — LLM agents build a living city in your browser. Three.js viewport, deterministic TS sim core, Ollama/DeepSeek agent pipeline.

Enterprise software for AI-powered business operations, intelligence, and automation.

Local-first Electron desktop workspace for building prompt-driven projects and coding-agent teams: prompt library, agent execution flows, CLI manager, project editor, design canvas. No backend, no account, your data never leaves the machine.

The design intelligence runtime for AI agents. 49 rules. Zero LLM. MCP-native. Give your agent eyes.

Self-hosted AI automation platform with unified LLM gateway, agents, workflows, and messaging bots. OpenClaw alternative.

A quiet Windows workspace for projects, tasks, and your AI agent. Local Markdown files, a command dock, and MCP. Built with Tauri and Svelte.

git for running agents — inspect, replay, share, fork AI coding sessions

deepseek-harness 启动器与运行环境管理器（Tauri 2 + Rust + React）

Mastra with built-in AI assistant

The browser layer for any AI agent - drives your real logged-in Chrome, learns workflows by watching you once, replays them with zero LLM tokens. MCP + HTTP + REST.

Mission-bounded authorization mesh for humans, workloads, and AI agents — deterministic policy enforcement across RBAC, ABAC, ReBAC, OAuth, Kubernetes, and agentic tool calls.

See where your Pi tokens and cost went: local session analytics for models, tools, subagents, skills, integrations, and history.

The missing image upload for AI agents on GitHub. One command attaches a screenshot to any issue or PR, hosted by GitHub itself.

Self-hosted MCP gateway: AI agents get audited, SSO-gated, read-only database access without ever holding a DB credential.

Port AI coding sessions and plans between Claude Code, OpenCode and Codex — without losing tool calls, timestamps or context.

Open-source orchestrator for Compound Engineering across AI coding agents. Turn ideas into verified work, preserve engineering context, and make each change improve the next.

Neovim plugin: run the real OpenCode TUI in a side panel and drive it over the v2 HTTP API.

TypeSafe Jev (System One) decision tools for Hermes Agent: jev_check / jev_route / jev_score / jev_evaluate

RevealUI, the agentic business runtime: users, content, products, payments, and agents. Core of RevealFleet.

🔍 Unrestricted, Self-Hosted Tavily Alternative for local AI agents. Features a 4-tier stealth scraper, 120ms HTTP-only search via a hybrid session pool, AI Overviews (SGE), and a machine-learning mouse trajectory solver.

3-tier hybrid memory plugin for OpenCode — compressed memory, atomic SQLite archive, session isolation, endless sessions via soft compaction

Build and audit an agentic OS layer by layer with Claude Code or Codex.

See which project each Claude Code, Codex and Cursor process belongs to, how long it has really been idle, and what it is costing you. Native macOS, no permissions required.

Pi coding agent extension for Fly.io Sprites: computers for agents. Routes Pi's native filesystem and shell tools into persistent Sprites, with checkpoints, services, worker pools, and RPC hosting.

Self-improvement feedback loop for AI agent skills — analyzes past sessions, drafts structured improvement proposals, and gates auto-apply behind an evaluation framework. Host-agnostic (Hermes, Claude Code, and any HostAdapter).

Agent Skill for implementing Google Stitch designs consistently with Codex, Claude Code and other coding agents.

Long-horizon project tracking for agents and their humans.

Open-source, self-hosted workspace where people and AI agents coordinate tasks, share reports, and collaborate under governed access.

A single-file 3D visualization of your AI agent ecosystem. Wireframe brain hull, neon neurons, live pulses, idle Earth-style rotation. No build step.

Omni-Agent: The Autonomous Business Intelligence Hub. AI-powered automation, real-time analytics, and enterprise-grade security.

Capgate — runtime authority boundary for autonomous AI agents: the agent proposes, the harness authorizes, the executor acts. Deny-by-default policy, non-transitive delegation, tamper-evident audit, MCP gateway, containerised isolation.

A complete legal assisstant for Hermes Agent — 1,432 skills across five layers. Apache-2.0.

Quy trình build app/SaaS cho người nocode bằng Claude Code skills + agents phản biện (Musk×Jobs). Idea→ship không rớt luồng.

Domain experts for Claude Code — persistent context, history, and institutional memory for your AI team

Bitcoin and macro market briefings for AI agents: trend signals, regimes, liquidity and ETF flows.

Laravel-native social publishing for Facebook, X, LinkedIn & TikTok with OAuth, queues, scheduling, retries, multi-account publishing & optional MCP.

MCP server for social media. Manage Instagram, WhatsApp, X, LinkedIn and 8 more DM inboxes plus scheduled posts from Claude, Cursor or ChatGPT.

Open-source governed-sandbox control plane for any workload — per-run identity, brokered credentials, layered egress, approvals, and an append-only audit. Coding agents are the flagship use. Apache-2.0, self-hosted.

E2BGateway is a unified gateway that enables the official E2B client to interoperate with multiple agent runtimes including agent-sandbox and OpenSandbox, delivering standardized sandbox APIs for AI agents.

Suna - Open Source Generalist AI Agent

local first tool to Stop rogue AI agents from deleting production. Runtime security layer that intercepts tool calls, blocks dangerous commands, and audits everything.

Claude Code pack for building award-style 3D/WebGL websites — 6 agents + 4 skills (Three.js, R3F, WebGPU, GSAP) + live demo.

Turn your AI coding agent sessions into a pixel-art RPG — every prompt, edit, and tool call earns XP and levels up your hero. Runs on Claude Code. 🦦

Accessibility runtime for streaming AI and agent interfaces—paced screen-reader announcements without rebuilding your UI.

Open source agentic programming environment for building, running, and inspecting AI workflows.

A local agent workspace for product design.

Security auditing CLI for AI-agent-authored code changesets: detects package hallucinations, indirect prompt injection, and AI-specific code defects before they hit production.

Meta-agent that composes paid skills — dinner + gift + transit in one request, with per-step receipts, powered by the suite

Framework that turns any website flow into an x402-paid API — declare a flow in YAML, Puppeteer executes, artifact returned in-response

Working Peerlytics and USDCtoFiat examples, templates, demos, and agent skills — by Galleon.

MCP servers for AI agents — Google Maps lead gen, SERP scraper, LinkedIn scraper, CAPTCHA solver, UAE real estate. Free, open-source, no API keys needed.

专利撰写助手 是一款面向研发、法务、知识产权岗位的 Windows 桌面应用。通过类 Cursor 的对话交互，按内置标准流程生成技术交底书八节、Mermaid 流程图与 Word 技术交底书；支持附件（图片、日志、PDF）、多模型路由与中英双语界面。

React-style MVU UI framework for C# desktop: pure C# components and hooks rendered to native WPF and Avalonia controls. Built for humans and AI agents.

Secure command center and gateway for agent-ready development environments

Claude Code skill (portable to any AI harness) that reviews UI/UX work through Luke Wroblewski's published design principles. Unofficial.

Split one booking across N wallets — the 6-person dinner problem: pooled x402 contributions with signed receipts and funded proofs

Standing orders for agents — mandate documents plus a client-side scheduler that pays per run and collects run reports

A fake town on testnet — mock restaurant, hotel, and store implementing the suite's exact contracts, for end-to-end agent purchase testing

Performance engineering and observability for AI agents.

Local markdown memory & cross-agent context engine for coding agents (MCP).

asobi — persistent knowledge graph CLI for humans and LLM agents

Guided setup wizard for new Agno projects.

Persistent intention infrastructure — goals, decisions, commitments, and strategic reasoning for AI agents

An open-source harness engineering system for Codex that turns product discovery into evidence-gated software delivery: PRDs, Stories, approved designs, implementation, testing, review, and pull requests.

Agentic LLM shell aimed at low cost GPUs

Build an LLM from scratch with transformers, attention, tokenization, dataset engineering, training, alignment, inference, multimodality, agents, and a runnable PyTorch mini language model.

SOC detection lab simulating AI-era threats, NHI abuse, prompt injection, and MCP attacks built with Splunk, Python, and MITRE ATT&CK mapping

Single-binary Rust microVM for AI-agent code execution. ~12 ms cold start, ~0.5 MiB Pss per fork.

Terminal workspace for AI agent sessions: one daemon owns PTY sessions in git worktrees, the TUI and CLI are clients over a unix socket.

Movie catalog concierge over TMDB — search, details, and recommendations per query, paid in USDC on Base or Solana via x402

The luck-robust benchmark for trading agents.

DXF and vector-PDF viewer for the web and AI agents — TypeScript parse→tessellate→render engine with a WebGL viewer, React/Vue/Svelte/web-component bindings, HTTP API, MCP server, and an in-chat viewer for ChatGPT & Claude (MCP Apps)

MCP server for Hound code search - expose regex-based code search to Claude Code and AI agents

Curated collection of system prompts from 35+ AI products (Claude, GPT, Gemini, Grok, Cursor, Devin, v0, etc.) — organized by category for reference and prompt engineering research

One identity for every coding agent you run. Marks each Claude Code, Antigravity, Codex or Gemini session with your label, the project, the machine and the moment.

Governance CLI for AI-native software delivery and AI coding agents. Enforces traceability: ADR → REQ → ROADMAP → kanban. Native support for Codex, Claude Code, Gemini CLI, Cursor, Copilot, Windsurf, and Amazon Q.

Agent skills for operating Azure Kubernetes Service (AKS): troubleshooting, cost, AKS Automatic readiness, GPU/inference, packet capture, and cluster design. The deep Day-2 AKS operator — complements the Azure Skills plugin; works with Claude Code, GitHub Copilot, and Azure SRE Agent.

Agent-to-agent price negotiation instruments — signed offers, counters, and agreements, each returned in-response

Evaluation harness for code-retrieval agents: mines ground truth from git commit history, then scores any agent on weighted F1 and end-to-end latency.

microservices.sh — scaffold and ship Cloudflare-native SaaS apps from verified modules (Workers, D1, SvelteKit, Hono).

Verified context for coding agents. Stop rediscovering your codebase.

🗼 An orchestrator skill for Claude Code: describe a task in plain language — Manara picks the stage, routes to the right skills, keeps guards blocking, and resumes across context clears.

VS Code extension that turns your TODO.md into an interactive Kanban board and runs autonomous Claude Code agent loops per task — markdown stays the source of truth.

Grow a fake company's entire sales history across CRM, calls, email and Slack that moves every week. Synthetic demo-data engine for B2B products.

Evidence-disciplined AI workflow skills for investor-style company research, diligence planning, versioned updates, and evidence review.

Local-first secrets manager for macOS. Encrypted vault, menu bar app, CLI, and an MCP server that loads secrets into your coding agent's shell instead of the chat. No cloud, no account, no subscription.

Transfer or resell held bookings between wallets — signed listings and reassigned booking tokens, all in-response