actiongate
Pre-execution safety layer for autonomous agent wallets. Risk scoring, transaction simulation, and policy enforcement via MCP.
README
ActionGate Examples
Integration examples for ActionGate — an x402 + Stripe payment proxy for REST APIs and MCP tools.
ActionGate ships with three built-in safety tools as its default example config:
| Tool | Description | Free Tier |
|---|---|---|
risk_score |
Evaluate the risk of a proposed agent action | 10 calls/day |
simulate |
Estimate costs and failure probability | 10 calls/day |
policy_gate |
Apply policy checks (allow / deny / allow-with-limits) | 5 calls/day |
Payment Options
ActionGate supports two payment methods:
- Free tier — no signup or API key needed. Rate-limited per tool.
- x402 micropayments — pay-per-call with USDC on Base. No account required.
- Stripe-funded API credits — buy prepaid credits at
/billing, receive anag_live_...API key, and calls deduct from your balance.
Framework Examples
| Framework | Directory | Description |
|---|---|---|
| LangChain | langchain/ |
ReAct agent using langchain-mcp-adapters |
| CrewAI | crewai/ |
Multi-agent crew with a treasury safety officer |
| Eliza (elizaOS) | eliza/ |
Character config with MCP plugin |
| OpenAI Agents SDK | openai-agents/ |
Agent with native MCP support |
SDKs
TypeScript
npm install actiongate-sdk
import { ActionGateClient } from "@actiongate/sdk";
const client = new ActionGateClient({
baseUrl: "https://api.actiongate.xyz",
apiKey: process.env.ACTIONGATE_API_KEY, // optional — omit for free tier or x402
});
const result = await client.policyGate({
actor: { actor_id: "agent_ops_02" },
action: {
action_type: "transfer",
network: "base",
asset_symbol: "USDC",
amount: "150000",
target: "0x1111111111111111111111111111111111111111",
},
policy: { policy_id: "treasury_default_v1" },
});
Python
pip install actiongate-sdk
from actiongate_sdk import ActionGateClient
client = ActionGateClient(
base_url="https://api.actiongate.xyz",
api_key=None, # optional — omit for free tier or x402
)
result = client.policy_gate({
"actor": {"actor_id": "agent_ops_02"},
"action": {
"action_type": "transfer",
"network": "base",
"asset_symbol": "USDC",
"amount": "150000",
"target": "0x1111111111111111111111111111111111111111",
},
"policy": {"policy_id": "treasury_default_v1"},
})
MCP Endpoint
All examples connect to the same endpoint:
https://api.actiongate.xyz/mcp
ActionGate uses Streamable HTTP transport. No authentication required for the free tier.
Remote MCP (any MCP-compatible client)
{
"mcpServers": {
"actiongate": {
"url": "https://api.actiongate.xyz/mcp"
}
}
}
Local stdio package
npm install -g actiongate-mcp
{
"mcpServers": {
"actiongate": {
"command": "npx",
"args": ["actiongate-mcp"],
"env": {
"PAYTO_ADDRESS": "0xYourWallet",
"ACTIONGATE_BASE_URL": "https://api.actiongate.xyz"
}
}
}
}
Registry namespace
xyz.actiongate.api/actiongate
MCP server quality
This public examples repo includes glama.json metadata for Glama indexing of the hosted ActionGate MCP endpoint. The production ActionGate implementation is maintained separately; this repo documents client integrations, proxy configuration, the hosted endpoint, maintainers, MIT license, tool annotations, input schemas, output schemas, and the intended pre-execution flow:
- Call
risk_scoreto classify the proposed action. - Call
simulateto estimate cost, failure probability, and side effects. - Call
policy_gateto return a final policy decision before execution.
All three default MCP tools are pre-execution checks. They do not submit transactions, move funds, or mutate policy packs.
BYO API (Proxy Config)
ActionGate can proxy any upstream API. Operators define tools in a proxy-config.json and ActionGate exposes them over REST (/proxy/:toolName) and MCP (/mcp) with x402 or Stripe billing attached.
See docs/proxy-config-guide.md for the full schema and walkthrough.
Quick example:
{
"serviceName": "My Paid API",
"serviceDescription": "ActionGate deployment for my upstream APIs.",
"defaultFreeTierCallsPerDay": 50,
"tools": [
{
"name": "summarize",
"description": "Summarize long text with my upstream model.",
"targetUrl": "https://example.com/v1/summarize",
"method": "POST",
"priceUsd": "0.25",
"freeTierLimit": 5,
"inputSchema": {
"type": "object",
"properties": { "text": { "type": "string" } },
"required": ["text"]
}
}
]
}
Pricing
The bundled safety tools use the following pricing:
| Tool | Price |
|---|---|
risk_score |
$0.02/call |
simulate |
$0.05–$1.00/call (tiered by action complexity) |
policy_gate |
$0.03/call |
Custom proxy deployments set their own pricing per tool in proxy-config.json.
Links
- ActionGate API
- Documentation
- Pricing
- OpenAPI Spec
- Proxy Config Guide
- Server Card
- Discovery JSON
- Billing Portal
License
MIT. See LICENSE.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.