Use Preloop to **onboard existing agents** with one command, and to **deploy event-driven agentic automations** with governed tools and budgets.
Preloop MCP server exists for a simple reason — assistants are far more useful when they can act on Preloop directly instead of describing what you should do. Use Preloop to onboard existing agents with one command, and to deploy event-driven agentic automations with governed tools and budgets.
Preloop is a single open-source platform that covers the five jobs teams otherwise buy from four different vendors:
Once Preloop is connected, these are the calls the assistant has available:
One — tap approvals** from your phone, watch, or desktop — for any onboarded agent's tool calls, including native Bash/Edit actions, not just MCP toolsPer — tool justification** — require (or optionally request) the agent to explain why a tool is being calledname — "deploy-approval"expression — "args.command.contains('deploy') && args.command.contains('production')"OpenAI — compatible** (/openai/v1/models, /openai/v1/chat/completions, /openai/v1/responses) and Anthropic-compatible (/anthropic/v1/messages) endpointsAllowed — model lists** per account, flow, API key, or managed agentRuntime — plugin dependent:** live OpenClaw and Hermes command delivery requires their native runtime plugin to honor preloop.control.control_ws_url, ownCapability — What it doesUpgrading — The installer detects the existing install and keeps its configuration (public URL, TLS setup, SMTP credentials, SECRET_KEY, database password)Being a remote server, there is no local install. You register the endpoint with your client, authorise it once, and the tools appear.
You will need 4 environment variables: PRELOOP_URL, SMTP_HOST, WEBAUTHN_RP_ID, SECRET_KEY. The server will not start without them, which is usually why the tools fail to appear on a first run. Keep credentials in your client's env block or a secrets manager rather than in a file you might commit.
This sits in the AI and media services group, where several servers overlap in what they claim to do but differ sharply once you actually set them up. Preloop's toolset — One, Per, name and 6 more — is a fair guide to whether it matches your workflow. It is maintained by preloop; worth a glance at recent repository activity before you build anything load-bearing on it.
This entry was verified against Preloop's own documentation before publication; SyncDev keeps the directory reviewed rather than auto-generated.
| Tool | What it does |
|---|---|
| One | tap approvals** from your phone, watch, or desktop — for any onboarded agent's tool calls, including native Bash/Edit actions, not just MCP tools. |
| Per | tool justification** — require (or optionally request) the agent to explain *why* a tool is being called. |
| name | "deploy-approval" |
| expression | "args.command.contains('deploy') && args.command.contains('production')" |
| OpenAI | compatible** (/openai/v1/models, /openai/v1/chat/completions, /openai/v1/responses) and **Anthropic-compatible** (/anthropic/v1/messages) endpoints with SSE streaming. |
| Allowed | model lists** per account, flow, API key, or managed agent. |
| Runtime | plugin dependent:** live OpenClaw and Hermes command delivery requires their native runtime plugin to honor preloop.control.control_ws_url, own reconnect/backoff and heartbeat/status loops, advertise capabilities, receiv |
| Capability | What it does |
| Upgrading | The installer detects the existing install and keeps its configuration (public URL, TLS setup, SMTP credentials, SECRET_KEY, database password), pulls the new images, **dumps the database to ~/.preloop-oss/backups/ first |
| Variable | Description | Required |
|---|---|---|
| PRELOOP_URL | Endpoint or connection string the server talks to. | Yes |
| SMTP_HOST | Endpoint or connection string the server talks to. | Optional |
| WEBAUTHN_RP_ID | Configuration value read at startup. | Optional |
| SECRET_KEY | Credential the server authenticates with. | Yes |
Build a programmable telecommunications stack for connecting telephony services with the Internet via a cloud-based utility.
Search built for AI, not humans — semantic web search that returns model-ready content, plus code context.
Answers, not links — delegate questions to Perplexity's search-grounded models and get cited responses back.
Give your assistant a voice — text-to-speech, voice cloning and audio tools from the ElevenLabs API.
Give your assistant a real code sandbox — isolated cloud VMs for actually running the code it writes.
The ML hub in your context window — search models, datasets, papers and run Spaces from the official server.