Screens prompts, output, URLs, EVM transactions and swaps. Heuristic. Free key at caelaralabs.com.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Screen a token and/or an EVM transaction: token safety score, decoded transaction risks (unlimited approvals, failed or contract-creating transactions), address-poisoning lookalike check (0.050 USDC per call)
EVM transaction hash (0x + 64 hex) to decode and screen
base|ethereum|arbitrum|optimism (default base)
Token contract (0x...) or Solana mint to screen
Address you intend to pay
Addresses you trust, to detect lookalike destinations
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http shield https://shield.caelaralabs.com/mcp
Screen a token and/or an EVM transaction: token safety score, decoded transaction risks (unlimited approvals, failed or contract-creating transactions), address-poisoning lookalike check (0.050 USDC per call)
| Argument | Type | Description |
|---|---|---|
| hash | string | EVM transaction hash (0x + 64 hex) to decode and screen |
| chain | string | base|ethereum|arbitrum|optimism (default base) |
| token | string | Token contract (0x...) or Solana mint to screen |
| destination | string | Address you intend to pay |
| known_addresses | array | Addresses you trust, to detect lookalike destinations |
{
"type": "object",
"required": [],
"properties": {
"hash": {
"type": "string",
"description": "EVM transaction hash (0x + 64 hex) to decode and screen"
},
"chain": {
"type": "string",
"description": "base|ethereum|arbitrum|optimism (default base)"
},
"token": {
"type": "string",
"description": "Token contract (0x...) or Solana mint to screen"
},
"destination": {
"type": "string",
"description": "Address you intend to pay"
},
"known_addresses": {
"type": "array",
"description": "Addresses you trust, to detect lookalike destinations"
}
}
}Fetch an https URL through an SSRF-guarded proxy and return a JSON envelope with the scanned body (HTTP 200 envelope, upstream status inside) (0.025 USDC per call)
| Argument | Type | Description |
|---|---|---|
| url | string | Target https URL (alternative to putting the URL in the path) |
| allow | string | Comma-separated hostnames or *.suffix patterns; when set the target must match |
| include_blocked | boolean | Return the body even when the content scan verdict is block (default false) |
{
"type": "object",
"required": [],
"properties": {
"url": {
"type": "string",
"description": "Target https URL (alternative to putting the URL in the path)"
},
"allow": {
"type": "string",
"description": "Comma-separated hostnames or *.suffix patterns; when set the target must match"
},
"include_blocked": {
"type": "boolean",
"description": "Return the body even when the content scan verdict is block (default false)"
}
}
}Screen untrusted text before it reaches a model: injection patterns, hidden characters, PII and secrets, with an allow/review/block verdict (0.015 USDC per call)
| Argument | Type | Description |
|---|---|---|
| text* | string | Untrusted input text (max 200000 characters) |
| redact | boolean | Return sanitized_text with PII and secrets masked (default true) |
{
"type": "object",
"required": [
"text"
],
"properties": {
"text": {
"type": "string",
"description": "Untrusted input text (max 200000 characters)"
},
"redact": {
"type": "boolean",
"description": "Return sanitized_text with PII and secrets masked (default true)"
}
}
}Check model output before it is used: secret and PII leakage, exfiltration links, active content, dangerous shell patterns, optional schema and grounding checks (0.015 USDC per call)
| Argument | Type | Description |
|---|---|---|
| output* | string | Raw model output (max 200000 characters) |
| schema | object | JSON Schema subset to enforce |
| source | string | Source text for a grounding check |
| system_prompt | string | Your system prompt, used only in this request to detect it being echoed back |
{
"type": "object",
"required": [
"output"
],
"properties": {
"output": {
"type": "string",
"description": "Raw model output (max 200000 characters)"
},
"schema": {
"type": "object",
"description": "JSON Schema subset to enforce"
},
"source": {
"type": "string",
"description": "Source text for a grounding check"
},
"system_prompt": {
"type": "string",
"description": "Your system prompt, used only in this request to detect it being echoed back"
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 20, 2026.
Takes a password, API key or token as an argument, so the secret passes through the model's context and chat logs.
Fix: Read credentials from the server's own configuration (environment, OAuth) instead of a tool argument.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.