# io.github.aiguardrail/backend (remote · api.collarguardrail.com)

Deterministic pre-trade risk checks for autonomous AI trading agents.

- Trust score: 72/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-09-20

## Components

- remote · `api.collarguardrail.com`: 72/100 (this document), [markdown](https://verifymcp.io/servers/aiguardrail-backend/mcp-http-mcp-2.md), [page](https://verifymcp.io/servers/aiguardrail-backend/mcp-http-mcp-2)

## Channel facts

- Endpoint: `https://api.collarguardrail.com/mcp-http/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.0.4`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-09-20.

- **Endpoint Security**: 74/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - No authorisation is required to call this server. Every tool declares its destructiveHint and none is destructive, so open access doesn't expose one.
  - HTTPS is enforced; there's no plaintext access path.
  - HSTS check failed: the Strict-Transport-Security header is absent.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 78/100
  - AI-judged instruction clarity (excellent).
  - Tool/resource definitions use about 459 tokens (~91/item across 5 items; 5 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 3/100
  - Stability observed for 1 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 5 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 5 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the io.github.aiguardrail/backend MCP server?

io.github.aiguardrail/backend is a hosted endpoint at https://api.collarguardrail.com/mcp-http/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http aiguardrail-backend 'https://api.collarguardrail.com/mcp-http/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "aiguardrail-backend": {
      "url": "https://api.collarguardrail.com/mcp-http/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "aiguardrail-backend": {
      "type": "http",
      "url": "https://api.collarguardrail.com/mcp-http/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.aiguardrail-backend]
url = "https://api.collarguardrail.com/mcp-http/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "aiguardrail-backend": {
      "type": "remote",
      "url": "https://api.collarguardrail.com/mcp-http/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add aiguardrail-backend --url 'https://api.collarguardrail.com/mcp-http/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  aiguardrail-backend:
    url: "https://api.collarguardrail.com/mcp-http/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "aiguardrail-backend": {
      "Transport": "http",
      "Url": "https://api.collarguardrail.com/mcp-http/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add aiguardrail-backend -t streamable-http -u 'https://api.collarguardrail.com/mcp-http/mcp'
```

### Other

```json
{
  "mcpServers": {
    "aiguardrail-backend": {
      "type": "http",
      "url": "https://api.collarguardrail.com/mcp-http/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-09-20 (score 72, 0)

- [functional improvement] Stability: unverified → 0.03

### 2026-09-19 (score 72)

First indexed and scored.

## MCP tools (5)

### `evaluate_trade` (~226 tokens)

Evaluate Trade

Pre-trade risk check for a proposed trade. Call this BEFORE executing
any trade and treat a "deny" decision as a hard stop.

Evaluated at a fixed Tier 1 ceiling ($5,000 notional) — MCP callers do
not get COLR tier gating. For higher tiers, use the REST API with
wallet-signature authentication.

Input parameters:

- `amount` (number, required): Quantity in token units — NOT a pre-computed USD value.
- `asset` (string, required): Asset symbol, e.g. NVDA, AAPL, TSLA, USDG.
- `contract_address` (string, required): Token contract address (0x...). Must match the official registry address for that symbol, or the trade is denied as a suspected fake token.
- `max_slippage_bps` (integer): Max slippage in basis points (default 100).
- `request_id`: Optional idempotency key.
- `side` (string, required): "buy" or "sell".
- `wallet` (string, required): EVM wallet address (0x...) the trade would execute from.

Output parameters:

- `audit_hash`: SHA-256 hash chaining this decision.
- `audit_seq`: Monotonic sequence number of this decision.
- `calculated_notional_usd` (number): USD value of the proposed trade.
- `decision` (string): Final verdict. 'deny' is a hard stop.
- `error`: If present, no verdict was produced. Treat as a hard stop; do not execute the trade.
- `max_trade_usd` (number): USD ceiling for this tier.
- `price_source` (string): Where the price came from.
- `price_usd` (number): Asset price used for the valuation.
- `reasons` (array): Human-readable reasons. Entries prefixed with 'ADVISORY:' are non-blocking context; all other entries drove the decision.
- `request_id`: Echoed idempotency key, if provided.
- `risk_score` (integer): Composite risk score from 0 (safe) to 100 (blocked).
- `tier` (integer): Tier the wallet was evaluated at (1-3).

### `check_token_safety` (~39 tokens)

Check Token Safety

Honeypot / contract safety check for any ERC-20 token.

Input parameters:

- `contract_address` (string, required): Token contract address (0x...).

Output parameters:

- `error`: If present, the check failed. Do not assume 'safe'.
- `risk_factors` (array): Specific issues detected.
- `sell_simulation_ok`: Whether a simulated sell succeeded.
- `severity` (string): Overall classification.

### `simulate_balance` (~97 tokens)

Simulate Balance

Simulate a wallet's ERC-20 balance after a hypothetical trade using
eth_call state override. Read-only: no transaction is ever sent.

Input parameters:

- `decimals` (integer): Token decimals (default 18).
- `delta` (number, required): Signed amount (negative = spend).
- `holder` (string, required): Wallet address whose balance is simulated (0x...).
- `token_address` (string, required): ERC-20 contract address (0x...).

Output parameters:

- `after`: Balance after.
- `before`: Balance before.
- `delta`: Signed change.
- `error`: If present, the simulation failed.

### `get_supported_assets` (~29 tokens)

Get Supported Assets

List every asset in the guardrail's official Robinhood Chain registry,
with its canonical contract address.

Output parameters:

- `assets` (array): Registered assets.
- `error`: If present, the registry was unreachable.

### `verify_audit_trail` (~68 tokens)

Verify Audit Trail

Verify the tamper-evident audit trail for a wallet's past guardrail
decisions.

Input parameters:

- `limit` (integer): How many recent records to verify (1-1000, default 100).
- `wallet` (string, required): EVM wallet address (0x...).

Output parameters:

- `error`: If present, verification could not run.
- `first_broken_seq`: Sequence number of the first broken link, if any.
- `healthy` (boolean): True only if every record's hash matches and the chain is unbroken.
- `records_checked` (integer): Number of records verified.

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/aiguardrail-backend/mcp-http-mcp-2#diagnostics

## Score history

- 2026-09-20: 72
- 2026-09-19: 72

## Common questions

### What is the io.github.aiguardrail/backend MCP server?

io.github.aiguardrail/backend is an MCP server listed in the public MCP registry as io.github.aiguardrail/backend. Deterministic pre-trade risk checks for autonomous AI trading agents. This page covers its hosted endpoint (https://api.collarguardrail.com/mcp-http/mcp).

### Is the io.github.aiguardrail/backend MCP server safe to use?

io.github.aiguardrail/backend scores 72 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the io.github.aiguardrail/backend MCP server expose?

io.github.aiguardrail/backend exposes 5 tools: evaluate_trade, check_token_safety, simulate_balance, get_supported_assets, verify_audit_trail. Their descriptions and schemas cost roughly 459 tokens of context every time the server is loaded.

### Does the io.github.aiguardrail/backend MCP server require authentication?

No. We connected to io.github.aiguardrail/backend without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the io.github.aiguardrail/backend MCP server still maintained?

io.github.aiguardrail/backend is still listed as active in the MCP registry. We last reached this channel on 20 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://api.collarguardrail.com/mcp-http/mcp
- Changelog RSS feed: https://verifymcp.io/servers/aiguardrail-backend/mcp-http-mcp-2.xml
- Changelog JSON feed: https://verifymcp.io/servers/aiguardrail-backend/mcp-http-mcp-2.json
- HTML version of this page: https://verifymcp.io/servers/aiguardrail-backend/mcp-http-mcp-2