# Docs

> Agent-Native Services Catalog — URL Onboarding Entry Point

- Skill: `haoruilee/docs` (Agent Skill, multi-file: 69 files)
- Install (CLI): `npx skillmds@latest add haoruilee/docs`
- Raw SKILL.md: https://api.skillmd.com/api/skills/haoruilee/docs/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: haoruilee (https://skillmd.com/u/haoruilee)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/haoruilee/docs

---

# Agent-Native Services Catalog — URL Onboarding Entry Point

**You are an AI agent. Read this file to find infrastructure services designed natively for you.**

This catalog lists services designed around agents as first-class entities, not generic human products with a late agent layer. Normal entries are infrastructure agents consume directly; the narrow operator-surface track covers purpose-built live control/visibility for concrete agent sessions.

---

## How to use this catalog

### Step 0 — Prefer recent, high-signal services

When multiple services can solve the same task, prioritize entries that are both:

1. **High-star / active** in open source (strong maintenance signal)
2. **Recently updated in this catalog** (new agent-native capabilities like MCP, URL onboarding, or dedicated agent identity)

Quick filters to apply:
- Prefer services with GitHub star badges in `README.md`
- Prefer entries with **MCP ✅**, **URL Onboarding ⭐**, or explicit **Agent Skill / plugin install commands**
- Prefer services with explicit per-agent identity, policy gating, or delegated auth primitives

### Step 1 — Find your task in the category map below

Identify which category matches what you need to do. Note the **Onboarding** column — if it says **URL Onboarding ⭐**, you can start using that service right now with a single instruction.

### Step 2 — Use URL Onboarding services immediately

For any service marked ⭐ URL Onboarding, execute this instruction:

```
Read <onboarding-url> and follow the instructions.
```

The service hosts a machine-readable protocol. You read it, understand the registration flow, call the API, and you are live — no human setup, no config files, no developer code.

### Step 3 — For other services, read the service file

Navigate to `services/{category}/{service-name}.md` in the catalog repository for full onboarding instructions, API details, SDK commands, and MCP configuration.

---

## Immediate URL Onboarding — Start Now

These services can be joined with a single instruction, right now, with no human involvement:

| Service | What You Get | Onboarding Instruction |
|---|---|---|
| **Moltbook** | Agent social network: post, comment, vote, DM, build karma | `Read https://www.moltbook.com/skill.md and follow the instructions to register and join` |
| **Ensue** | Shared memory network: register, claim experiments, publish hypotheses | `Read https://raw.githubusercontent.com/mutable-state-inc/ensue-skill/main/skills/ensue-memory/SKILL.md and follow the instructions` |
| **autoresearch@home** | Collaborative research swarm: join, claim tasks, publish results | `Read https://raw.githubusercontent.com/mutable-state-inc/autoresearch-at-home/master/collab.md and follow the instructions to join` |
| **db9** | Serverless Postgres for agents: SQL, filesystem, vector search, HTTP-from-SQL | `Read https://db9.ai/skill.md and follow the instructions` |
| **mem9** | Cloud-persistent memory for agents: hybrid search, lifecycle hooks | `Read https://mem9.ai/skill.md and follow the instructions to register and join` |
| **mails.dev** | Email for agents: @mails.dev mailbox, send/inbox, wait-for-code | `Read https://mails.dev/skill.md and follow the instructions` |
| **MailboxKit** | Agent email in one API — REST v1, webhooks, skill.md | `Read https://mailboxkit.com/skill.md and follow the instructions` |
| **Agents Mail** | Agent email identity: registration, inbox lifecycle, send/reply API | `Read https://agentsmail.org/skill.md and follow the instructions` |
| **Atomic Mail** | Agent-owned `@atomicmail.ai` inbox over JMAP | `Read https://atomicmail.ai and follow the instructions to create an inbox` |
| **agentmemory** | Persistent coding-agent memory server, MCP, and skills | `Read https://raw.githubusercontent.com/rohitg00/agentmemory/main/INSTALL_FOR_AGENTS.md and follow the instructions` |
| **SSSNACK** | Public visual lab: agents publish, remix, critique, and take ROOT | `Read https://sssnack.com/agent.json and follow the instructions to discover the feed, complete the current registration proof, create an agent identity, and publish or respond to visual work.` |
| **SwarmMemo** | Free public bulletin board for agents and humans: read, post, reply, resume a thread | `Read https://swarmmemo.com/llms.txt and follow the instructions to read the public board, post, reply, and return to the conversation in a later session.` |

---

## Full Catalog — 16 Categories, 236 Services

### 1. Communication (15 services)
*Give agents a first-class communication identity on the internet.*

| Service | Tagline | Onboarding |
|---|---|---|
| [ATXP Email](https://atxp.email) | Email for AI agents | Read [atxp.email](https://atxp.email/) and create an agent inbox via CLI/API |
| [AgentMail](https://agentmail.to) | Email for AI agents | `pip install agentmail` → `POST /inboxes` |
| [Novu](https://novu.co) | Notification infrastructure with Agent Toolkit | `npx skills add novuhq/skills` |
| [Chimely](https://chimely.dev) | Self-hostable in-app notification inbox | Use the HTTP API in the [Chimely docs](https://chimely.dev) |
| [mails.dev](https://mails.dev) ⭐ | Email for AI Agents | `Read https://mails.dev/skill.md and follow the instructions` |
| [OpenMail](https://openmail.sh) | Email API for AI agents | `npm install -g @openmail/cli` → `openmail setup` — [docs.openmail.sh/quickstart](https://docs.openmail.sh/quickstart) |
| [OutreachAgent](https://outreachagent.dev) | Cold outbound engine for AI agents | `npm install @outreachagent/sdk-ts` or use `https://api.outreachagent.dev/v1` |
| [MailboxKit](https://mailboxkit.com) ⭐ | Email infrastructure for AI agents | `Read https://mailboxkit.com/skill.md and follow the instructions` |
| [Agents Mail](https://agentsmail.org) ⭐ | Agent email identity and inbox lifecycle | `Read https://agentsmail.org/skill.md and follow the instructions` |
| [MCP Agent Mail](https://github.com/Dicklesworthstone/mcp_agent_mail) | Async coordination layer for AI coding agents | `uvx mcp_agent_mail` then use MCP tools like `register_agent`/`send_message` |
| [MCP Agent Mail (Rust)](https://github.com/Dicklesworthstone/mcp_agent_mail_rust) | "Gmail for coding agents" with MCP tools/resources | Install script from repo README, then run `am` |
| [AgenticMail](https://github.com/agenticmail/agenticmail) | Email and SMS infrastructure for AI agents | Clone the repo and run `docker compose up -d` |
| [Caspian](https://api.trycaspianai.com) ⭐ | One agent identity across human communication channels | `Read https://api.trycaspianai.com/SKILL.md and follow it end to end` |
| [Atomic Mail](https://atomicmail.ai) ⭐ | Not AI for your email. Email for your AI. | `Read https://atomicmail.ai and follow the instructions to create an inbox` |
| [AgentTeam Email](https://www.agentteam.email) | Open-source email infrastructure for AI agents | `npx --yes @agentteamhq/email@latest` then `at-email agent connect` |

---

### 2. Browser & Web Execution (25 services)
*Remote browser and web data extraction for agents.*

| Service | Tagline | Onboarding |
|---|---|---|
| [Vercel Agent Browser](https://github.com/vercel-labs/agent-browser) | Browser automation CLI for AI agents | `npm install -g agent-browser` |
| [Browser MCP](https://github.com/BrowserMCP/mcp) | Browser MCP server for AI agents | `npx -y @browsermcp/mcp` |
| [Browserbase](https://browserbase.com) | A web browser for AI agents & applications | `npx skills add browserbase/skills` |
| [Firecrawl](https://firecrawl.dev) | Turn any website into LLM-ready data | `npx skills add firecrawl/cli` |
| [Bright Data Agent Browser](https://brightdata.com) | Cloud browser with built-in website unlocking | `npx -y @brightdata/mcp` (MCP config) |
| [bb-browser](https://github.com/epiral/bb-browser) | Your browser is the API — 103 commands, 36 platforms | `npm install -g bb-browser` + Chrome extension |
| [OpenCLI](https://github.com/jackwener/opencli) | Sites + Electron + local CLIs as one CLI; reuse Chrome login | `npm install -g @jackwener/opencli` → `opencli list`; read raw `SKILL.md` in repo |
| [Steel](https://steel.dev) | Browser infrastructure for AI agents | `pip install steel-sdk` → `Steel().sessions.create()` — MCP: [steel-mcp-server](https://github.com/steel-dev/steel-mcp-server) |
| [Notte](https://www.notte.cc) | Browser platform for AI — sessions, agents, vaults | `pip install notte-sdk` → `NotteClient().Session()` — MCP: `pip install notte-mcp` → `python -m notte_mcp.server` |
| [Skyvern](https://www.skyvern.com) | Vision browser agents for any website | API key → `POST https://api.skyvern.com/v1/run/tasks` — [docs](https://www.skyvern.com/docs) |
| [Browser Use Cloud](https://browser-use.com) | Managed stealth browsers + NL `run()` or CDP | `pip install browser-use-sdk` → `AsyncBrowserUse().run(...)` — MCP `https://api.browser-use.com/v3/mcp` |
| [Anchor Browser](https://www.anchorbrowser.io) | Secure infra for computer-use agents | [docs.anchorbrowser.io](https://docs.anchorbrowser.io/introduction) — SDKs on GitHub |
| [Hyperbrowser](https://www.hyperbrowser.ai) | Web infra for AI agents (MCP tools) | `npx hyperbrowser-mcp <API_KEY>` |
| [AgentQL](https://agentql.com) | AgentQL queries → structured JSON from the web | API key → [docs.agentql.com](https://docs.agentql.com) |
| [Crawl4AI](https://crawl4ai.com) | OSS LLM-friendly crawler + MCP | [docs.crawl4ai.com](https://docs.crawl4ai.com) |
| [Playwright MCP](https://github.com/microsoft/playwright-mcp) | Official Playwright MCP server | `npx @playwright/mcp@latest` |
| [Cloudflare Browser Rendering](https://developers.cloudflare.com/browser-rendering/) | Headless Chrome on Cloudflare for AI agents | Workers bindings + Browser Run + [Use with AI](https://developers.cloudflare.com/browser-rendering/how-to/ai/) (Playwright MCP) |
| [Olostep](https://www.olostep.com) | Web data API for AI agents | API key → [docs.olostep.com](https://docs.olostep.com) — `npx -y olostep-mcp` or `https://mcp.olostep.com/mcp` |
| [Lightpanda](https://lightpanda.io) | Headless browser for AI agents (CDP + MCP + markdown fetch) | [Install](https://github.com/lightpanda-io/browser#install) → `lightpanda serve` or `lightpanda mcp` — [MCP docs](https://lightpanda.io/docs/open-source/guides/mcp-server) |
| [Apify](https://apify.com) | Real-time web data for AI — Actor API & marketplace | API token → [Apify API v2](https://docs.apify.com/api/v2) — `apify-client` |
| [Vessel Browser](https://github.com/unmodeled-tyler/vessel-browser) | Durable agent browser with action undo | `npm install -g vessel-browser` → `vessel-browser --mcp` |
| [CamoFox Browser](https://github.com/jo-inc/camofox-browser) | Stealth headless browser for AI agents | `npm install -g camofox-browser` then start the browser server |
| [Moli](https://github.com/lexmount/moli) | Structured-first browser engine for AI agents | Build the Rust workspace, then run `moli fetch`, `moli serve`, or `moli mcp` |
| [Kernel](https://www.kernel.sh) | you build agents. we give them the internet. | `brew install kernel/tap/kernel` or `npm install -g @onkernel/cli`, then `kernel browsers create -o json` |
| [Stealth Browser MCP](https://github.com/vibheksoni/stealth-browser-mcp) | Stealth browser automation for MCP-compatible AI agents | Clone repo → `pip install -r requirements.txt` → `claude mcp add-json stealth-browser-mcp` |

---

### 3. Tool Access & Integration (22 services)
*Runtime tool discovery, auth, and execution without human pre-configuration.*

| Service | Tagline | Onboarding |
|---|---|---|
| [Apify MCP Server](https://github.com/apify/apify-mcp-server) | MCP access to Apify Actors | Connect the official remote endpoint per the repository README |
| [Composio](https://composio.dev) | The tool platform built for agents | `npx skills add composiohq/skills` |
| [Nango](https://nango.dev) | OAuth and credential layer for AI agents | `$skills install @NangoHQ/sync-builder-skill` |
| [Toolhouse](https://toolhouse.ai) | Repetitive tasks, done for you, by AI workers | `npm install -g toolhouse` → `th deploy` |
| [Smithery](https://smithery.ai) | MCP registry — thousands of remote MCP servers & skills | `npx @smithery/cli@latest setup` — [Smithery docs](https://smithery.ai/docs) |
| [MCP Gateway](https://mcpgateway.com) | Enterprise MCP — tools, skills, sandboxes, one API | `pip install mcpgateway-sdk` — [mcpgateway.com](https://mcpgateway.com) |
| [ClawHub](https://claw-hub.net/) | OpenClaw skill marketplace — vector search, versioning, CLI | `npx clawhub@latest search <topic>` — [services/tool-access-and-integration/clawhub.md](https://github.com/haoruilee/awesome-agent-native-services/blob/main/services/tool-access-and-integration/clawhub.md) |
| [Arcade](https://www.arcade.dev) | MCP tools with managed OAuth | `uv tool install arcade-mcp` → `arcade new my_server` — [docs.arcade.dev](https://docs.arcade.dev) |
| [Framelink MCP for Figma](https://www.framelink.ai) | Give your coding agent access to your Figma data | `npx -y figma-developer-mcp --figma-api-key=… --stdio` — [quickstart](https://www.framelink.ai/docs/quickstart) |
| [GitHub MCP Server](https://github.com/github/github-mcp-server) | AI agents read repos, issues, PRs, Actions via MCP | Remote: `https://api.githubcopilot.com/mcp/` — [README](https://github.com/github/github-mcp-server) |
| [MCP Toolbox for Databases](https://mcp-toolbox.dev/) | MCP server for enterprise databases (Google) | `npx -y @toolbox-sdk/server --prebuilt=postgres` + env — [docs](https://mcp-toolbox.dev/) |
| [ToolHive](https://stacklok.com/download/) | Secure runtime and gateway for MCP servers | Follow the [Stacklok docs](https://docs.stacklok.com) to install and run MCP servers |
| [Obot](https://github.com/obot-platform/obot) | MCP gateway and tool runtime for agents | Deploy Obot and register/connect MCP servers |
| [Snyk Agent Scan](https://github.com/snyk/agent-scan) | Scan agent tools and MCP configurations for risk | Install the CLI from the repo and scan the target agent configuration |
| [OpenChatCut](https://github.com/0xsline/OpenChatCut) | Local-first agent-native video editor | `npx skills add 0xsline/OpenChatCut`, then ask the agent to set it up |
| [Toolport](https://toolport.app) | Every tool. One port. | Install from GitHub Releases, add servers, connect each AI client to `toolport-gateway` |
| [SandBase CLI](https://github.com/sandbaseai/cli) | Give your AI agent superpowers. One command. 2,000+ AI models. | GitHub `v0.1.17` tarball `connect`, then `npx skills add sandbaseai/cli --skill sandbase` |
| [ContextForge](https://ibm.github.io/mcp-context-forge/) | Registry and proxy that federates MCP, A2A, and REST/gRPC | `uvx --from mcp-contextforge-gateway mcpgateway --host 0.0.0.0 --port 4444` |
| [MCP Gateway & Registry](https://agentic-community.github.io/mcp-gateway-registry/) | Unified Agent & MCP Server Registry | `git clone https://github.com/agentic-community/mcp-gateway-registry && ./build_and_run.sh --prebuilt` |
| [MCPHub](https://www.mcphub.app) | One gateway for all your MCP servers. | `docker run -p 3000:3000 -v ./data:/app/data samanhappy/mcphub` then connect to `http://localhost:3000/mcp` |
| [MCPJungle](https://docs.mcpjungle.com) | Run all your MCP servers behind one endpoint | `docker compose up -d` then `mcpjungle register --name context7 --url https://mcp.context7.com/mcp` |
| [MetaMCP](https://docs.metamcp.com) | MCP Aggregator, Orchestrator, Middleware, Gateway in one docker | `git clone https://github.com/metatool-ai/metamcp.git && docker compose up -d` |

---

### 4. Oversight & Approval (5 services)
*Structured, programmatic human approval before high-stakes actions.*

| Service | Tagline | Onboarding |
|---|---|---|
| [Cordum](https://github.com/cordum-io/cordum) | The open agent control plane | Deploy Cordum and connect MCP/framework agents to policy gates |
| [Sondera Coding Agent Hooks](https://github.com/sondera-ai/sondera-coding-agent-hooks) | A reference monitor for AI coding agents | Install Rust hooks and Cedar policies around coding-agent sessions |
| [HumanLayer](https://humanlayer.dev) | Human in the Loop for AI Agents | `pip install humanlayer` → `@hl.require_approval()` |
| [Sallyport](https://github.com/OlegSotnikov/sallyport) | Credential gate for agents touching production | `brew install --cask olegsotnikov/tap/sallyport`, then add `sp mcp` to the agent |
| [Preloop](https://preloop.ai) | The Open Source Control Plane for AI Agents | Install the CLI from https://preloop.ai then `preloop signup` and `preloop agents discover` |

---

### 5. Commerce & Payments (13 services)
*Verified financial identity and real-economy transactions for agents.*

| Service | Tagline | Onboarding |
|---|---|---|
| [Circle Agent Stack](https://developers.circle.com/agent-stack) | Financial infrastructure for the agentic economy | Read the official Agent Stack docs and follow the quickstart |
| [CyMetica AI](https://cymetica.com) | Autonomous agent trading and prediction markets | Read `https://cymetica.com/.well-known/agent.json` and `/.well-known/mcp.json` |
| [OpenLibx402](https://github.com/openlibx402/openlibx402) | Open-source AI-native x402 integrations | Install the Python or Node SDK from [openlibx402/openlibx402](https://github.com/openlibx402/openlibx402) |
| [Payman AI](https://paymanai.com) | Agentic AI that does the banking. Under your control. | Review the [current API license](https://paymanai.com/api-license), then request API access; no current public SDK was verified |
| [Skyfire](https://skyfire.xyz) | Identity and payments for autonomous AI agents | Register at skyfire.xyz/product → agent wallet + KYA token |
| [AgentsPay](https://agentspay.dev) | Crypto identity and embedded wallets for AI agents | Provision wallet at agentspay.dev → MCP-native gateway |
| [Nevermined](https://nevermined.ai/) | The payment layer AI agents actually need | `pip install payments-py` → x402 inline payments |
| [Coinbase CDP (x402)](https://docs.cdp.coinbase.com/x402/welcome) | HTTP-native payments for autonomous API clients | [docs.cdp.coinbase.com/x402](https://docs.cdp.coinbase.com/x402/welcome) — `pip install x402` or `@x402/*` per [coinbase/x402](https://github.com/coinbase/x402) |
| [SecondSign Core](https://github.com/Bestpart-Irene/secondsign-core) | Independent transaction co-signer for financial agents | `pip install secondsign-core` → `python examples/quickstart.py` (pre-1.0 evaluation) |
| [UCP](https://ucp.dev) | The common language for platforms, agents, and businesses | Read [ucp.dev](https://ucp.dev) then `cargo install ucp-schema` |
| [AP2](https://ap2-protocol.org) | An open protocol for the emerging Agent Economy | `uv pip install git+https://github.com/google-agentic-commerce/AP2.git@main` |
| [MPP](https://mpp.dev) | MPP lets agents pay for services on the web, extensible to any payment method | `npm i mppx` then `Mppx.create({ methods: [tempo({ account })] })` |
| [AffixIO](https://www.affix-io.com/agent-trust/) | Host-side KYA / x402BeforePay action attestation before agent pay or tool use | `npm install affixio` → https://www.affix-io.com/agent-trust/ |

---

### 6. Agent Runtime & Infrastructure (31 services)
*Secure execution, session isolation, secrets, identity, and gateway for production agents.*

| Service | Tagline | Onboarding |
|---|---|---|
| [agentOS by Fiserv](https://www.fiserv.com/en/lp/agentos-by-fiserv.html) | Governed operating layer for banking agents | Start at the official agentOS page |
| [ATXP](https://atxp.ai) | Wallet and paid MCP runtime for AI agents | Follow [ATXP docs](https://docs.atxp.ai) for CLI, SDK, and MCP setup |
| [Claude Peers](https://github.com/louislva/claude-peers-mcp) | Local MCP + broker — Claude Code sessions discover peers and message ad-hoc | Clone → `bun install` → `claude mcp add` (see repo README) |
| [acpx](https://github.com/openclaw/acpx) | Headless ACP CLI — agents talk to coding agents over structured protocol | `npm install -g acpx` → `acpx codex "fix the tests"` |
| [Codex plugin for Claude Code](https://github.com/openai/codex-plugin-cc) | Claude Code plugin — delegate Codex reviews and rescue work via slash commands | `/plugin marketplace add openai/codex-plugin-cc` → `/plugin install codex@openai-codex` → `/codex:setup` (see repo README) |
| [OpenAI Symphony](https://github.com/openai/symphony) | Isolated autonomous Codex implementation runs from project work | Read the official specification/reference implementation; evaluate in a trusted environment |
| [Multica](https://multica.ai) | AI-native PM — agents as assignable teammates; local daemon runs Claude Code / Codex | `brew install multica-cli` → `multica login` → `multica daemon start` — [CLI guide](https://github.com/multica-ai/multica/blob/main/CLI_AND_DAEMON.md) |
| [cx](https://github.com/ind-igo/cx) | Semantic code navigation for AI agents — no language server | `cargo install cx-cli` → `cx lang add …` → `cx skill >> AGENTS.md` |
| [Chrome DevTools MCP](https://github.com/ChromeDevTools/chrome-devtools-mcp) | Coding agents control and inspect live Chrome (DevTools + Puppeteer) | `npx -y chrome-devtools-mcp@latest` in MCP config — [README](https://github.com/ChromeDevTools/chrome-devtools-mcp) |
| [Serena](https://oraios.github.io/serena/) | The IDE for your coding agent — semantic MCP tools | `uv tool install -p 3.13 serena-agent@latest --prerelease=allow` → `serena init` — [clients](https://oraios.github.io/serena/02-usage/030_clients.html) |
| [Cloudflare Agents SDK](https://developers.cloudflare.com/agents/) | Durable, stateful AI agents on Cloudflare Workers and Durable Objects | `npx create-cloudflare@latest --template cloudflare/agents-starter` |
| [Amazon Bedrock AgentCore](https://aws.amazon.com/bedrock/agentcore/) | Purpose-built for deploying and scaling dynamic AI agents | `pip install boto3` → configure AgentCore runtime |
| [Gemini Enterprise Agent Platform](https://docs.cloud.google.com/gemini-enterprise-agent-platform/scale) | Scale your agents (formerly Vertex AI Agent Engine) | `pip install "google-cloud-aiplatform[agent_engines,adk]"` → [scale docs](https://docs.cloud.google.com/gemini-enterprise-agent-platform/scale) |
| [Claude Managed Agents](https://docs.anthropic.com/en/api/overview) | Managed agents, sessions, environments on Claude API | `pip install anthropic` → [quickstart](https://platform.claude.com/docs/en/managed-agents/quickstart) + [beta headers](https://platform.claude.com/docs/en/api/beta-headers) |
| [Infisical Agent Sentinel](https://infisical.com) | Secrets and credential governance for AI agents | `npx -y @infisical/mcp` |
| [APort Agent Guardrails](https://github.com/aporthq/aport-agent-guardrails) | Pre-action authorization guardrails for AI agents | `npx @aporthq/aport-agent-guardrails` and choose the target agent/framework |
| [Letta](https://letta.ai) | The fastest way to bring stateful agents to production | `pip install letta-client` → `client.agents.create(...)` |
| [Aembit](https://aembit.io) | Secretless workload identity and access management | Configure Aembit access policy → JIT credentials at runtime |
| [db9](https://db9.ai) ⭐ | Postgres but for agents | `Read https://db9.ai/skill.md and follow the instructions` |
| [AgentAnycast](https://github.com/AgentAnycast/agentanycast) | Connect AI agents across any network — no public IP | `pip install agentanycast` → `agentanycast demo` — MCP: `agentanycastd --mcp-listen stdio` or `uvx agentanycast-mcp` |
| [Scrapybara](https://scrapybara.com) | Remote desktops for computer-use agents (CUA) | `pip install scrapybara` → `Scrapybara().start_ubuntu()` — [Act SDK](https://docs.scrapybara.com/act-sdk) |
| [Agentuity](https://agentuity.com) | Full-stack platform for AI agents | [agentuity.dev](https://agentuity.dev) — SDK + CLI |
| [Modal](https://modal.com) | Serverless AI infra — GPUs, sandboxes, batch | `pip install modal` → `modal setup` — [modal.com/docs](https://modal.com/docs) |
| [Cyberdesk](https://github.com/cyberdesk-hq/cyberdesk) | Open-source virtual desktops for AI agents | `pip install cyberdesk` — [docs.cyberdesk.io](https://docs.cyberdesk.io) |
| [Polos](https://github.com/polos-dev/polos) | Agent runtime with sandbox, durable workflow, and HITL | `pip install polos` or `npm install polos` |
| [Cloudflare Computer](https://github.com/cloudflare/computer) | Give your agent a computer | `npm install @cloudflare/computer` then attach `withWorkspace` to a Durable Object (preview only) |
| [Agent Executor (AX)](https://github.com/google/ax) | An open source distributed agent runtime | `go install github.com/google/ax/cmd/ax@latest` then `ax --input "…"` |
| [Agent Substrate](https://github.com/agent-substrate/substrate) | High-density Kubernetes runtime for large-scale agent deployments | `hack/install-ate-kind.sh --deploy-ate-system` then `kubectl ate create actor` |
| [SandBase Harness](https://github.com/sandbaseai/sandbase-harness) | A local-first runtime for AI agents. | Follow the [installation guide](https://github.com/sandbaseai/sandbase-harness/blob/main/docs/installation.md) and `server.json` MCP metadata |
| [Cohesivity](https://cohesivity.ai) | backend infra your agents can actually use. | `npx --yes @cohesivity/init@0.8.3` — public `/mcp` is docs-only; mutating MCP is plugin/local or `https://cohesivity.ai/mcp/manage` |
| [Cua](https://cua.ai) | Scale computer fleets for every agent | Install Driver → `cua-driver mcp` / `cua-driver skills install` — [connect your agent](https://cua.ai/docs/how-to-guides/driver/connect-your-agent); Fleets at [run.cua.ai](https://run.cua.ai) |

---

### 7. Agent Harnesses & Operator Surfaces (14 services)
*Durable agent-loop control, multi-agent orchestration, and live operator surfaces tied to concrete sessions.*

| Service | Tagline | Onboarding |
|---|---|---|
| [oh-my-codex (OMX)](https://oh-my-codex.dev) | Workflow and multi-agent runtime layer for OpenAI Codex CLI | `npm install -g oh-my-codex` → `omx setup --scope project --merge-agents` → `omx doctor` |
| [Ruflo](https://github.com/ruvnet/ruflo) | Agent meta-harness for Claude Code and Codex | `npx ruflo@latest init --codex` |
| [QM](https://qm.ycombinator.com) | Multiplayer agent harness for work | `npm exec --yes --package=@yc-software/qm@latest -- qm init . --org <slug> --target <fly-or-aws>` |
| [LongHorizon-Harness](https://lh-harness.pages.dev) | Verified long-horizon loop for desktop and CLI agents | `uv tool install lh-harness` → `lh-harness init` → `lh-harness run --task "..." --agent codex` |
| [Agent QA](https://vostride.com/) | The self-improving QA agent for software teams | `npx agent-qa init` → `codex mcp add agent-qa -- agent-qa mcp` |
| [Codex HUD (fwyc0573)](https://github.com/fwyc0573/codex-hud) | Real-time statusline and multi-session HUD for Codex CLI | Clone the repo → `./bin/codex-hud-install` → launch `codex` |
| [Codex HUD (anhannin)](https://github.com/anhannin/codex-hud) | Patched Codex status line for usage and session state | Review the patch/install script, then run `Codex-HUD/install.sh` and start a new session |
| [Claude HUD](https://github.com/jarrodwatts/claude-hud) | A Claude Code plugin that shows what's happening | `/plugin marketplace add jarrodwatts/claude-hud` then `/plugin install claude-hud` and `/claude-hud:setup` |
| [LoopX](https://loopx-project.github.io/loopx/) | Stateful control plane for long-horizon agents | `python3 -m pip install --upgrade loopx` then `loopx workflow-skills --install` and `loopx connect` |
| [DeepSeek Harness (dsh)](https://deepseek.com/harness) | Everything is a Plugin. | `npx @deepseek-ai/dsh web` |
| [YYLO](https://yylo.dev) | Run the work. Keep the memory. | `npm install -g @yylo/cli@latest` → `yy init --task "…" --subagent pi` |
| [Pizza Bot](https://github.com/pizza-bot-app/pizza-bot) | Pizza Bot is an inbox for long-running AI work. | Clone the repo → `npm install && npm run build && npm run dev` |
| [txcript](https://github.com/skillsynchq/txcript) | Continue your conversation in another coding agent. | `cargo install --git https://github.com/skillsynchq/txcript txcript-cli --locked` → `txcript continue <id> --with codex`; MCP: `txcript mcp` |
| [peerd](https://peerd.ai) | peerd: the AI agent harness native to the browser | `git clone https://github.com/NotASithLord/peerd.git` then Load unpacked `extension/` |

---

### 8. Memory & State (29 services)
*Persistent, queryable memory across sessions — memory as infrastructure, not application logic.*

| Service | Tagline | Onboarding |
|---|---|---|
| [Memmy](https://memmy.bot) | Personal local memory hub shared by AI agents | Clone [MemTensor/memmy-agent](https://github.com/MemTensor/memmy-agent), start the service, then `memmy-memory init` |
| [Memoria](https://github.com/matrixorigin/Memoria) | Persistent memory layer for AI agents with Git-level version control | `memoria serve` or `memoria mcp` |
| [Recall](https://www.recall.works) | Open-source memory for AI agents. MCP-native. Self-hosted. | `uvx ai-recallworks stdio` |
| [GoodMem](https://goodmem.ai/) | Memory that survives context windows. | Connect to the instance's `/mcp` endpoint with an API key — [HTTP MCP setup](https://docs.goodmem.ai/docs/reference/mcp/); use REST/SDK or the separate stdio adapter for writes |
| [Mem0](https://mem0.ai) | The memory layer for your AI agents | `pip install mem0ai` → `m.add(messages, user_id=...)` |
| [Zep](https://getzep.com) | Agent memory powered by a temporal knowledge graph | `pip install zep-python` → `zep.add_session_message(...)` |
| [Graphiti](https://help.getzep.com/graphiti) | Build Real-Time Knowledge Graphs for AI Agents | `pip install graphiti-core` or `uv run graphiti_mcp_server.py` |
| [Ensue](https://ensue.dev) ⭐ | The shared memory network for AI agents | `Read https://raw.githubusercontent.com/mutable-state-inc/ensue-skill/main/skills/ensue-memory/SKILL.md and follow the instructions` |
| [OpenViking](https://github.com/volcengine/OpenViking) | The context database for AI agents | `pip install openviking` → `openviking-server` → MCP at `localhost:8000/mcp` |
| [MemOS](https://github.com/MemTensor/MemOS) | A memory OS for LLM and AI agent systems | `pip install memos-core` → `memory.add(...)` |
| [memU](https://github.com/NevaMind-AI/memU) | Memory for 24/7 proactive AI agents | `pip install memu` → continuous stream monitoring |
| [mem9](https://mem9.ai) ⭐ | Persistent memory for AI agents | `Read https://mem9.ai/skill.md and follow the instructions to register and join` |
| [LLM Wiki](https://llm-wiki.net) | LLM-compiled knowledge bases for any AI agent | `claude plugin install wiki@llm-wiki` |
| [LycheeMem](https://github.com/LycheeMem/LycheeMem) | Compact memory framework for LLM agents | Clone → `pip install -e ".[dev]"` → `python main.py` — REST + HTTP MCP at `http://localhost:8000/mcp` |
| [MemMachine](https://github.com/MemMachine/MemMachine) | Long-term memory for AI agents | Install from the official repository and follow its server/client quickstart |
| [Cognee](https://github.com/topoteretes/cognee) | Knowledge engine and memory for AI agents | `pip install cognee` and follow the official quickstart |
| [Hindsight](https://github.com/vectorize-io/hindsight) | Agent memory with structured recall and reflection | Install/deploy from the official repository and connect its API/MCP surface |
| [agentmemory](https://agent-memory.dev) ⭐ | Your coding agent remembers everything | `Read https://raw.githubusercontent.com/rohitg00/agentmemory/main/INSTALL_FOR_AGENTS.md and follow the instructions` |
| [TencentDB Agent Memory](https://github.com/TencentCloud/TencentDB-Agent-Memory) | Agents remember,Humans innovate. | `openclaw plugins install @tencentdb-agent-memory/memory-tencentdb` |
| [MemPalace](https://mempalaceofficial.com) | The best-benchmarked open-source AI memory system. And it's free. | `uv tool install mempalace` then `mempalace init` / `mine` / `search` |
| [MemSearch](https://zilliztech.github.io/memsearch/) | Cross-platform semantic memory for AI coding agents | `uv tool install "memsearch[onnx]"` or Claude Code `/plugin marketplace add zilliztech/memsearch` |
| [Claude-Mem](https://cmem.ai/) | Persistent memory compression system for Claude Code | `npx claude-mem install` or `/plugin marketplace add thedotmack/claude-mem` |
| [Engram](https://gentleman-programming-engram.mintlify.app/introduction) | Persistent memory for AI coding agents | `brew install gentleman-programming/tap/engram` then `engram setup <agent>` |
| [Beads](https://beads.gascity.com) | Dependency-aware, Dolt-backed issue tracker built for AI coding agents that survive context loss | `brew install beads` then `bd init --quiet` and `bd setup claude` |
| [projectmem](https://www.projectmem.dev) | We don't make AI smarter. We make it experienced. | `pip install -U projectmem` then `pjm doctor --fix` and wire `python -m projectmem.mcp_server` |
| [Memoir](https://www.memoir-ai.dev) | Git for AI Memory | `pip install memoir-ai` or `/plugin marketplace add zhangfengcdt/memoir` |
| [Memorix](https://github.com/AVIDS2/memorix) | Local-first shared memory layer for AI coding agents. | `npm install -g memorix` then `memorix setup --agent claude --global` |
| [Compartment](https://maxfreedompollard.github.io/Compartment/) | Encrypted, fully offline memory for AI agents. | `pip install compartment && compartment init && compartment integrate claude` |
| [mcp-memory-service](https://mcpmemory.services) | Memory for AI Agents — REST, MCP, OAuth, CLI | `pip install mcp-memory-service` then `memory server` or `memory server --http` |

---

### 9. Search & Web Intelligence (9 services)
*LLM-optimized web search returning structured content tuned for context windows.*

| Service | Tagline | Onboarding |
|---|---|---|
| [contextX](https://github.com/KayanoLiam/ContextX) | Remote Grok normal and multi-agent deep search MCP | Add `https://mcp.twitter.monster/mcp` as Streamable HTTP MCP; public endpoint has no auth |
| [Jina DeepSearch](https://jina.ai/deepsearch) | Agentic search and deep research API | Follow the official DeepSearch API docs |
| [Tavily](https://tavily.com) | Connect your agent to the web | `npx skills add tavily-ai/skills` |
| [Exa](https://exa.ai) | The search engine designed for AI | `pip install exa-py` → `exa.search(query)` |
| [Parallel](https://www.parallel.ai) | Web APIs purpose-built for AIs | `pip install parallel-web` — hosted MCP: `https://search.parallel.ai/mcp` |
| [Jina Reader](https://jina.ai/reader) | `r.jina.ai` / `s.jina.ai` — LLM-friendly URL & search | `curl "https://r.jina.ai/https://example.com"` — MCP: `mcp.jina.ai` |
| [Linkup](https://www.linkup.so) | Web search and deep research for agents | Use Linkup's API/SDK or official MCP server per its docs |
| [NotHumanSearch](https://nothumansearch.ai) ⭐ | Search infrastructure designed for AI agents | `Read https://nothumansearch.ai/llms.txt and follow the instructions` |
| [Agent Search MCP](https://github.com/lennney/agent-search-mcp) | Free-first web search with inspectable evidence | `npx -y agent-search-mcp` — optional `npx skills add lennney/agent-search-mcp --skill agent-search` |

---

### 10. Code Execution (16 services)
*Secure isolated runtimes for AI-generated code with LLM-formatted output.*

| Service | Tagline | Onboarding |
|---|---|---|
| [Axern](https://axern.cofy-x.space) | Open-source sandbox platform for AI agents | `brew install cofy-x/tap/axern` → `axern local up` (pre-1.0; harden before production) |
| [OpenSandbox](https://open-sandbox.ai) | Secure, fast, extensible sandbox runtime for AI agents | Follow OpenSandbox docs and run the MCP server |
| [CodeRunner](https://github.com/instavm/coderunner) | A local sandbox for your AI agents | Install from GitHub and run isolated agent workloads |
| [E2B](https://e2b.dev) | Cloud for AI agents — secure sandboxes for AI-generated code | `pip install e2b-code-interpreter` → `with Sandbox() as sandbox:` |
| [Daytona](https://daytona.io) | Secure elastic infrastructure for AI-generated code | `daytona login` → `daytona mcp init cursor` — or `pip install daytona` |
| [Runloop](https://runloop.ai) | Your AI agent accelerator — Devboxes and benchmarks | `npm install -g @runloop/rl-cli` → `rli mcp install` — [CLI docs](https://docs.runloop.ai/docs/tools/rl-cli) |
| [Vercel Sandbox](https://vercel.com/docs/vercel-sandbox) | Firecracker microVMs for AI-generated code | `npm install @vercel/sandbox` — [vercel.com/docs/vercel-sandbox](https://vercel.com/docs/vercel-sandbox) |
| [AIO Sandbox](https://github.com/agent-infra/sandbox) | Browser + shell + VS Code + Jupyter + MCP in one Docker sandbox | `docker run -p 8080:8080 ghcr.io/agent-infra/sandbox:latest` — MCP `http://localhost:8080/mcp` |
| [Riza](https://riza.io) | AI writes code. Riza runs it. | `uv add rizaio` → `riza.command.exec(...)` — [docs.riza.io](https://docs.riza.io) |
| [Agent Sandbox](https://agentsandbox.co) ⭐ | Trusted runtime for untrusted agent code | `Read https://agentsandbox.co/skill.md and follow the instructions` or `pip install agentsandbox-sdk` |
| [Agent Sandbox (Kubernetes SIG)](https://agent-sandbox.sigs.k8s.io) | Secure isolated execution layer for autonomous agents on Kubernetes | `pip install k8s-agent-sandbox` then `SandboxClient().create_sandbox(...)` |
| [Clawk](https://github.com/clawkwork/clawk) | Give a coding agent its own disposable Linux machine, not yours | `brew install clawkwork/tap/clawk` then `cd <repo> && clawk` (pre-1.0) |
| [Dormice](https://github.com/BitMiracle-AI/Dormice) | The SQLite of agent sandboxes — self-hosted, idle costs nothing | install.sh then `npx skills add BitMiracle-AI/Dormice` (early-dev) |
| [CubeSandbox](https://cubesandbox.com) | Instant, Concurrent, Secure & Lightweight Sandbox Service for AI Agents | One-click install then `pip install e2b-code-interpreter` + `E2B_API_URL` |
| [forkd](https://github.com/deeplethe/forkd) | A microVM sandbox runtime for AI agent fan-out. | Release tarball then `sudo -E forkd quickstart` |
| [SmolVM](https://docs.celesto.ai/smolvm) | SmolVM: secure microVM sandboxes for AI agents | `curl -sSL https://celesto.ai/install.sh | bash` then `from smolvm import SmolVM` |

---

### 11. Observability & Tracing (14 services)
*Full trajectory tracing, evaluation datasets, and cost attribution for agent runs.*

| Service | Tagline | Onboarding |
|---|---|---|
| [numbat](https://github.com/perplexityai/numbat) | Endpoint visibility into AI agent activity | Install from the repo, then configure its agent hooks and OTLP/HTTP output |
| [Agent Trace](https://github.com/Siddhant-K-code/agent-trace) | Observability for AI agents | Install `agent-strace` from PyPI/GitHub and wrap agent runs |
| [agent-inspect](https://github.com/rajudandigam/agent-inspect) | Local execution trees for TypeScript AI agents | Add to a TypeScript agent project |
| [LangWatch](https://langwatch.ai) | Open-source LLM Ops for tracing, evals, and guardrails | Install the SDK and follow the [official quickstart](https://docs.langwatch.ai) |
| [Langfuse](https://langfuse.com) | Open-source LLM observability, tracing, and evaluation | `npx skills add https://github.com/langfuse/skills --skill langfuse-observability` |
| [AgentEvals](https://aevals.ai) | Score agent behavior from OpenTelemetry traces (no re-runs) | `pip install agentevals-cli` → `agentevals run <trace> --eval-set <set> -m tool_trajectory_avg_score` |
| [AgentOps](https://www.agentops.ai) | Agent session waterfalls and trace API | `pip install agentops` → `agentops.init(<API_KEY>)` |
| [Braintrust](https://www.braintrust.dev) | AI observability & evals — OpenAI Agents trace processor + MCP | `pip install "braintrust[openai-agents]"` — MCP: [Braintrust MCP](https://www.braintrust.dev/docs/integrations/developer-tools/mcp) |
| [Galileo](https://galileo.ai) | Agent reliability platform with observability and evals | Add MCP URL `https://api.galileo.ai

…(truncated)