MCP trading gateway for AI agents: IRL pre-trade policy, sealed rationale, verifiable fills
- ✓Open-source license (MIT)
- ✓Actively maintained (<30d)
- ✓Clear description
- ✓Documented (README)
git clone https://github.com/macropulse-lab/irl-gateway && cp irl-gateway/*.md ~/.claude/agents/Subagents overview
# IRL Gateway
<!-- mcp-name: io.github.macropulse-lab/irl-gateway -->
**Give your AI agent a trading account it can't misuse, and a record of every decision it can't rewrite.**
IRL Gateway is an [MCP](https://modelcontextprotocol.io) server that sits between an AI agent (Claude, ChatGPT, or your own) and an exchange account. Every order the agent places goes through the [IRL Engine](https://irl.macropulse.live):
1. **Policy before execution.** IRL checks the order against the agent's mandate (active status, notional cap, allowed assets and venues) before anything reaches the exchange. Out of mandate means no order.
2. **The rationale is sealed.** The agent must say why it is trading. The gateway hashes that rationale together with the trade inputs and seals the hash into IRL's tamper-evident trace, anchored daily to Bitcoin. The plaintext stays in your local journal.
3. **Intent is reconciled with the fill.** After the exchange fills the order, IRL compares what was authorized with what executed and records `MATCHED` or `DIVERGENT`.
When something goes wrong, you can prove what the agent was allowed to do, what it said it was doing, and what actually happened.
```
AI agent ── MCP ──> irl-gateway ──> IRL: authorize (policy + sealed rationale)
│
├──────> exchange: market order (client id = sealed intent)
│
└──────> IRL: bind fill -> MATCHED / DIVERGENT
```
## Tools
| Tool | What it does |
| --- | --- |
| `execute_trade(symbol, side, rationale, quantity \| notional)` | The only tool that moves money. Spot market order through authorize → place → bind. Returns `filled`, `denied`, `blocked` or `failed`, with the IRL `trace_id` and verdict. |
| `get_policy()` | The agent's mandate as IRL enforces it, plus the local kill-switch state. |
| `get_quote(symbol)` | Last price on the gateway's venue. |
| `get_balances()` | Free balances (paper or exchange). |
| `get_trace(trace_id)` | IRL's sealed record of one trade. |
| `list_recent_trades(limit)` | Local journal: rationale, context hash, trace id and outcome per trade. |
Behaviour the agent can rely on:
- **Fail closed.** If IRL is unreachable or denies the intent, no order is sent.
- **Kill switch.** Create the file `~/.irl-gateway/KILL` and every trade is refused before IRL is even called. Delete it to resume.
- **No silent fills.** If the exchange fills but the IRL bind fails, the result still reports the fill and flags it for reconciliation.
## Quick start (paper trading)
You need an IRL server and an agent registered on it. Paper trading is the default: fills are simulated at live public Binance prices, and no exchange keys are needed.
```bash
pip install irl-gateway # or run it without installing: uvx irl-gateway
```
Register the agent once, with its mandate:
```bash
curl -X POST "$IRL_BASE_URL/irl/agents" -H "Authorization: Bearer $IRL_API_TOKEN" \
-H "Content-Type: application/json" -d '{
"name": "my-claude-trader",
"model_hash_hex": "<sha256 of your agent config>",
"max_notional": 100,
"allowed_assets": ["BTC/USDT", "ETH/USDT"],
"allowed_venues": ["paper-binance"]
}'
```
Then add the gateway to your MCP client, for example Claude Code or Claude Desktop:
```json
{
"mcpServers": {
"irl-gateway": {
"command": "uvx",
"args": ["irl-gateway"],
"env": {
"IRL_BASE_URL": "https://irl.example.com",
"IRL_API_TOKEN": "…",
"IRL_AGENT_ID": "<agent_id from registration>",
"IRL_MODEL_HASH": "<the same model_hash_hex>",
"AGENT_MODEL_ID": "claude-opus-5-5",
"PAPER_BALANCES": "USDT=1000"
}
}
}
}
```
Ask the agent to check `get_policy`, then trade.
## Configuration
| Variable | Default | Meaning |
| --- | --- | --- |
| `IRL_BASE_URL`, `IRL_API_TOKEN` | required | IRL server and bearer token |
| `IRL_AGENT_ID`, `IRL_MODEL_HASH` | required | The registered agent and its model hash |
| `AGENT_MODEL_ID` | `unspecified-model` | Model name sealed into each trace (the agent can override it per trade) |
| `AGENT_CONFIG_CHECKSUM` | `none` | Optional checksum of the agent's configuration, sealed into each trace |
| `IRL_L2_MODE` | `off` | `regime` if your IRL server requires Layer 2 regime binding |
| `GATEWAY_BROKER` | `paper` | `paper` or `exchange` |
| `EXCHANGE_ID` | `binance` | Any ccxt exchange id; also the price source for paper trading |
| `EXCHANGE_API_KEY`, `EXCHANGE_API_SECRET` | | Required for `exchange` |
| `EXCHANGE_TESTNET` | `true` | Use the exchange's testnet |
| `PAPER_BALANCES` | `USDT=1000` | Starting paper balances (used only until `paper_state.json` exists; the paper account then persists across restarts) |
| `IRL_GATEWAY_HOME` | `~/.irl-gateway` | Journal (`journal.jsonl`), kill switch (`KILL`) and paper account (`paper_state.json`) location |
The venue IRL sees is the exchange id (`binance`), or `paper-<exchange>` for paper trading, so a mandate can allow paper trading while denying the real account.
## How the rationale is sealed
For each trade the gateway builds a context of the rationale, symbol, side, quantity, reference price, venue, model id and client order id. It hashes that context as canonical JSON (sorted keys, no whitespace) with SHA-256 and sends the hash to IRL as `prompt_version = "ctx-sha256:<hex>"`, which IRL seals into the trace's `reasoning_hash`.
The journal stores the full context next to its hash, so anyone holding a journal line can recompute the hash and match it to the sealed trace. IRL itself never sees the rationale's text.
## Development
```bash
python -m venv .venv && .venv/bin/pip install -e ".[dev]"
pytest --cov=irl_gateway
ruff check src tests && black --check src tests && isort --check-only src tests && mypy src
```
## Status
Early (0.1). Spot market orders only. Paper trading and ccxt exchanges are supported; Alpaca is next. Not investment advice, and no strategy is included: the gateway controls and records what your agent does, it does not decide.
MIT licensed.
What people ask about irl-gateway
What is macropulse-lab/irl-gateway?
+
macropulse-lab/irl-gateway is subagents for the Claude AI ecosystem. MCP trading gateway for AI agents: IRL pre-trade policy, sealed rationale, verifiable fills It has 0 GitHub stars and its last recorded update is dated 2026-10-05.
How do I install irl-gateway?
+
You can install irl-gateway by cloning the repository (https://github.com/macropulse-lab/irl-gateway) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is macropulse-lab/irl-gateway safe to use?
+
Our security agent has analyzed macropulse-lab/irl-gateway and assigned a Trust Score of 87/100 (tier: Trusted). See the full breakdown of passed checks and flags on this page.
Who maintains macropulse-lab/irl-gateway?
+
macropulse-lab/irl-gateway is maintained by macropulse-lab. The last recorded GitHub activity is dated 2026-10-05, with 0 open issues.
Are there alternatives to irl-gateway?
+
Yes. On ClaudeWave you can browse similar subagents at /categories/agents, sorted by popularity or recent activity.
Deploy irl-gateway to your cloud
Ship this repo to production in minutes. Each platform spins up its own environment with editable env vars.
Maintain this repo? Add a badge to your README
Drop the badge into your GitHub README to show it's tracked on ClaudeWave. Each badge links back to this page and reflects the live Trust Score.
[](https://claudewave.com/repo/macropulse-lab-irl-gateway)<a href="https://claudewave.com/repo/macropulse-lab-irl-gateway"><img src="https://claudewave.com/api/badge/macropulse-lab-irl-gateway" alt="Featured on ClaudeWave: macropulse-lab/irl-gateway" width="320" height="64" /></a>More Subagents
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
The agent that grows with you
Java 面试 & 后端通用面试指南,覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
The agent engineering platform.