aegis-pay 0.3.4

Semantic Payment Guardrail for AI Agents

Project Aegis (AgentPay)

Note: This is the PyPI published documentation. For the full architecture diagrams, real UI screenshots, and the Chinese translation (繁體中文版), please visit the GitHub Repository.

Project Aegis is a payment guardrail and one-time flow protocol specifically designed for Agentic AI (e.g., OpenClaw, NemoClaw, Claude Code, OpenHands). It enables agents to handle financial transactions safely without risking unlimited exposure of human-controlled credit cards.

1. The Problem

When Agentic AI encounters a paywall (e.g., domain registration, API credits, compute scaling) during an automated workflow, it is often forced to stop and wait for human intervention. However, providing a physical credit card directly to an agent introduces a "trust crisis": hallucinations or infinite loops could lead to the card being drained.

2. Dual Architecture

Project Aegis is designed with a "Dual Architecture" vision to scale from open-source local experiments to enterprise-grade AI production pipelines.

1. Hacker Edition (BYOC + DOM Injection)

Built for open-source frameworks like OpenClaw and NemoClaw. The agent never receives the true credit card number—it only sees a masked version (`****-4242`). When the agent successfully navigates to a checkout paywall, the Aegis `AegisBrowserInjector` attaches to the active Chromium browser via the Chrome DevTools Protocol (CDP). It precisely traverses all cross-origin iframes (like Stripe Elements) and injects the real credentials deep into the DOM form elements, delivering 100% protection against prompt injection or hallucination-driven extractions. Bring Your Own Card (BYOC) locally with absolute peace of mind.

2. Enterprise Edition (Stripe Issuing)

The "North Star" for the broader Agentic SaaS ecosystem. Proving that Aegis has the enterprise-grade extensibility required for the real world, it seamlessly connects to verified financial infrastructure. Perfect for platforms building "Agentic Visa" services that programmatically issue real, single-use, burner virtual credit cards (VCCs) via the Stripe API for cloud-hosted AI fleets.

---

3. Ecosystem Position: Aegis + Browser Agents = Unstoppable

Modern agentic workflows require two complementary capabilities. Aegis does one, and does it exceptionally well.

🎯 What Aegis Is — and Isn't

Aegis is the agent's financial brain and safe vault. It is responsible for:

• ✅ Evaluating whether a purchase should happen (semantic guardrails)
• ✅ Enforcing hard budget limits (daily cap, per-transaction cap)
• ✅ Issuing one-time virtual cards so real credentials are never exposed
• ✅ Maintaining a full audit trail of every payment attempt

Aegis does NOT:

• ❌ Navigate websites or interact with DOM elements
• ❌ Solve CAPTCHAs or bypass bot-detection systems
• ❌ Fill out forms or click "Submit" on behalf of the agent

That's the browser agent's job.

🤝 The Handshake: How Aegis and Browser Agents Work Together

The real power emerges when Aegis is paired with a browser automation agent (e.g., OpenHands, browser-use, Skyvern). The workflow is a clean division of labor:

1. [Browser Agent]  Navigates to a site, scrapes product info, reaches checkout.
        │
        │  (Hit a paywall / payment form)
        ▼
2. [Browser Agent → Aegis MCP]  Calls request_virtual_card(amount, vendor, reasoning)
        │
        │  (Aegis evaluates: budget OK? vendor approved? no hallucination?)
        ▼
3. [Aegis]  Issues a one-time virtual card (Stripe mode) or mock card (dev mode)
            Returns masked card number to agent. Full card injected only via
            trusted local execution environment — never into the LLM's context.
        │
        ▼
4. [Browser Agent]  Uses the approved credentials to complete the checkout form.
        │
        ▼
5. [The Vault]  Dashboard logs the transaction. Card is immediately burned.

🌐 Supported Browser Agent Integrations

Browser Agent	Integration Method	Guide
OpenHands	MCP Tool Call	Quick Start §5
OpenClaw + browser-use	MCP Tool Call	Quick Start §5
NemoClaw (sandboxed)	MCP Tool Call inside sandbox	Quick Start §5
Custom Playwright / Selenium	Python SDK AegisClient	Integration Guide
Skyvern / browser-use	Python SDK middleware	Integration Guide

See docs/INTEGRATION_GUIDE.md for end-to-end code examples including Playwright injection and System Prompt templates.

---

4. Installation

# Core only (keyword guardrail + mock provider, zero external dependencies)
pip install aegis-pay

# With LLM-based guardrails (supports OpenAI, Ollama, vLLM, OpenRouter)
pip install aegis-pay[llm]

# With Stripe virtual card issuing
pip install aegis-pay[stripe]

# With LangChain integration
pip install aegis-pay[langchain]

# Full installation (all features)
pip install aegis-pay[all]

5. Quick Start for OpenClaw / NemoClaw / Claude Code / OpenHands

If you're using OpenClaw, NemoClaw, Claude Code, OpenHands, or any MCP-compatible agentic framework, you can get Aegis running in under 2 minutes:

Step 1: Install & Start MCP Server

# Clone the repo
git clone https://github.com/TPEmist/Project-Aegis.git
cd Project-Aegis

# Install dependencies
uv sync --all-extras

# Start the MCP server
uv run python -m aegis.mcp_server

Step 2: Connect to Your Agent

OpenClaw:

# Register Aegis as an MCP tool in OpenClaw
openclaw mcp add aegis -- uv run python -m aegis.mcp_server

# Or add to your OpenClaw MCP config file (~/.openclaw/mcp_servers.json)

{
  "aegis": {
    "command": "uv",
    "args": ["run", "python", "-m", "aegis.mcp_server"],
    "cwd": "/path/to/Project-Aegis",
    "env": {
      "AEGIS_ALLOWED_CATEGORIES": "[\"aws\", \"cloudflare\", \"openai\"]",
      "AEGIS_MAX_PER_TX": "100.0",
      "AEGIS_MAX_DAILY": "500.0"
    }
  }
}

NemoClaw (NVIDIA sandbox):

NemoClaw wraps OpenClaw agents in a secure sandbox. Configure Aegis inside your NemoClaw sandbox by editing the MCP config within the sandbox environment:

# Connect to your NemoClaw sandbox
nemoclaw my-assistant connect

# Inside the sandbox, register the Aegis MCP server
openclaw mcp add aegis -- uv run python -m aegis.mcp_server

Note: NemoClaw restricts file access. Make sure Project-Aegis is cloned inside /sandbox/ so the agent can access it. The aegis_state.db will be created in the sandbox's writable directory.

Claude Code:

claude mcp add aegis -- uv run python -m aegis.mcp_server

OpenHands: Add to your MCP configuration:

{
  "mcpServers": {
    "aegis": {
      "command": "uv",
      "args": ["run", "python", "-m", "aegis.mcp_server"],
      "cwd": "/path/to/Project-Aegis"
    }
  }
}

Step 3: Configure Your Policy (Environment Variables)

export AEGIS_ALLOWED_CATEGORIES='["aws", "cloudflare", "openai", "github"]'
export AEGIS_MAX_PER_TX=100.0        # Max $100 per single transaction
export AEGIS_MAX_DAILY=500.0         # Max $500 per day total
export AEGIS_BLOCK_LOOPS=true        # Block hallucination/retry loops
# Optional: export AEGIS_STRIPE_KEY=sk_live_... (see §8 for Stripe setup)

Step 4: Use It

Your agent now has access to the request_virtual_card tool. When it encounters a paywall:

Agent: "I need to purchase an API key from AWS for $15 to continue."
[Tool Call] request_virtual_card(amount=15.0, vendor="AWS", reasoning="Need API key for deployment")
[Aegis] ✅ Payment approved. Card Issued: ****4242, Expiry: 12/25, Amount: 15.0
Agent: "Purchase successful, continuing workflow."

If the agent hallucinates or tries to overspend:

Agent: "Let me retry buying compute... the previous attempt failed again."
[Tool Call] request_virtual_card(amount=50.0, vendor="AWS", reasoning="failed again, retry loop")
[Aegis] ❌ Payment rejected. Reason: Hallucination or infinite loop detected in reasoning

---

6. Core Components

🛡️ The Vault

A local visualization console powered by Streamlit and SQLite (aegis_state.db). The Vault allows humans to:

• Monitor all issued seals and agent spending activity in real-time.
• Monitor global budget utilization.
• Audit rejection logs from semantic guardrails.

📜 The Seal

Virtual, single-use payment credentials with built-in enforcement:

• Daily Budget Limit Enforcement: Automatically blocks any request that would exceed the predefined daily spending cap.
• Burn-after-use Interception: Ensures that once a virtual card is used, it is immediately invalidated, preventing replay attacks or unauthorized recurring charges.

🧠 Semantic Guardrails

Aegis provides two modes of intent evaluation to prevent agents from wasting funds:

1. Fast Keyword-based Interception (Default): Uses the GuardrailEngine to immediately block requests containing keywords associated with loops or hallucinations (e.g., "retry", "failed again", "ignore previous"). Zero dependencies, zero cost.
2. LLM-based Guardrail Engine: Powered by the LLMGuardrailEngine, this mode performs deep semantic analysis of the agent's reasoning to detect unrelated purchases or logical inconsistencies. Supports any OpenAI-compatible endpoint — including local models via Ollama/vLLM, or cloud providers like OpenAI and OpenRouter.

7. Security Statement

Security is a first-class citizen in Aegis. The SDK masks card numbers by default (e.g., ****-****-****-4242) when returning authorization results to the agent. This prevents sensitive payment information from leaking into agent chat logs, model context windows, or persistent logs, ensuring that only the execution environment handles the raw credentials.

8. The Vault Dashboard

The Vault is your real-time monitoring console for all agent payment activity.

Starting the Dashboard

cd Project-Aegis
uv run streamlit run dashboard/app.py
# Dashboard opens at http://localhost:8501

Dashboard Layout

Section	Description
Sidebar: Max Daily Budget slider	Adjust the displayed budget cap for visualization (does not affect backend policy — backend policy is configured via env vars or SDK)
Today's Spending	Total amount spent by agents today
Remaining Budget	How much budget is left for the day
Budget Utilization	Visual progress bar showing spend %
💳 Issued Seals & Activity	Full table of all payment attempts (approved + rejected) with seal ID, amount, vendor, status, and timestamp
🚫 Rejected Summary	Filtered view showing only rejected/blocked attempts for quick auditing

Tips

• Click Refresh Data in the sidebar to pull latest activity from the database.
• The dashboard reads from aegis_state.db — the same database the SDK writes to. Keep both running simultaneously for live monitoring.
• Each row in the table corresponds to a single request_virtual_card call from an agent.

---

9. Python SDK Quickstart

Integrate Aegis into your custom Python or LangChain workflows in just a few lines:

from aegis.client import AegisClient
from aegis.providers.stripe_mock import MockStripeProvider
from aegis.core.models import GuardrailPolicy

# Define your safety policy
policy = GuardrailPolicy(
    allowed_categories=["API", "Cloud", "SaaS"], 
    max_amount_per_tx=50.0, 
    max_daily_budget=200.0,
    block_hallucination_loops=True
)

# Initialize the client with keyword-only guardrails (default)
client = AegisClient(
    provider=MockStripeProvider(), 
    policy=policy,
    db_path="aegis_state.db"
)

# Or use LLM-based guardrails with a local model (e.g., Ollama)
from aegis.engine.llm_guardrails import LLMGuardrailEngine

llm_engine = LLMGuardrailEngine(
    base_url="http://localhost:11434/v1",  # Ollama endpoint
    model="llama3.2",
    use_json_mode=False
)
client = AegisClient(
    provider=MockStripeProvider(),
    policy=policy,
    engine=llm_engine
)

# Use with LangChain Tool
from aegis.tools.langchain import AegisPaymentTool
tool = AegisPaymentTool(client=client, agent_id="agent-01")

Supported LLM Providers

Provider	base_url	model
OpenAI (default)	(not needed)	gpt-4o-mini
Ollama (local)	http://localhost:11434/v1	llama3.2
vLLM / LM Studio	http://localhost:8000/v1	Your model name
OpenRouter	https://openrouter.ai/api/v1	anthropic/claude-3-haiku
Any OpenAI-compatible	Your endpoint URL	Your model name

---

10. Payment Providers: Stripe vs Mock

Without Stripe (Default — Mock Provider)

By default, Aegis uses the MockStripeProvider which simulates virtual card issuance. This is perfect for:

• Development and testing — no real money involved
• Demo and evaluation — see the full flow without any API keys
• Hackathons — get a working prototype in minutes

Mock cards are fully functional within the Aegis system (budget tracking, burn-after-use, guardrails all work), but they are not real payment instruments.

from aegis.providers.stripe_mock import MockStripeProvider

client = AegisClient(
    provider=MockStripeProvider(),  # No API key needed
    policy=policy
)

With Real Stripe Issuing

To issue real virtual credit cards through Stripe Issuing:

Prerequisites:

1. A Stripe account with Issuing enabled (requires application approval)
2. Your Stripe secret key (sk_live_... or sk_test_...)

Option A: Via Environment Variable (for MCP Server)

export AEGIS_STRIPE_KEY=sk_live_your_stripe_key_here
uv run python -m aegis.mcp_server
# The MCP server will automatically use StripeIssuingProvider

Option B: Via Python SDK

from aegis.providers.stripe_real import StripeIssuingProvider

client = AegisClient(
    provider=StripeIssuingProvider(api_key="sk_live_your_stripe_key_here"),
    policy=policy
)

What Stripe Issuing does:

• Creates a real Stripe Cardholder (Aegis Agent)
• Issues a virtual card with a spending limit matching the approved amount
• Returns masked card details (last 4 digits only) to the agent
• All Stripe errors are caught and returned as rejection reasons

Note: Stripe Issuing is a premium Stripe product that requires approval. For most development and demo use cases, the Mock provider is sufficient.