Ein MCP-Server (Machine Communication Protocol), der als Gateway für die Wise-API dient und einen vereinfachten Zugriff auf die Empfängerfunktionen von Wise bietet.
Auflisten aller Empfänger aus Ihrem Wise-Konto über eine einfache MCP-Ressource
Automatische Handhabung von Authentifizierung und Profilauswahl
Nutzt die Wise Sandbox-API für Entwicklung und Tests
Verfügbar als Docker-Image für eine einfache Integration
Related MCP server: poke-bank
Python 3.12 oder höher (nur bei direkter Installation)
uv Paketmanager (nur bei direkter Installation)
Wise API-Token
Docker (bei Verwendung des Docker-Images)
https://wise.com/your-account/integrations-and-tools/api-tokens
Erstellen Sie hier ein neues Token.
Klonen Sie dieses Repository:
git clone repo-url
cd jc-wise-mcpRichten Sie die Umgebung ein:
cp .env.example .env
# Edit .env to add your Wise API tokenInstallieren Sie die Abhängigkeiten mit uv:
uv venv
uv pip install -e .Sie können ein Docker-Image erstellen:
docker build -t mcp-wise .Fügen Sie es zu Claude Code hinzu, indem Sie es in Ihre .mcp.json eintragen:
{
"mcpServers": {
"mcp-wise": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"--init",
"-e", "WISE_API_TOKEN=your_api_token_here",
"-e", "WISE_IS_SANDBOX=true",
"mcp-wise:latest"
]
}
}
}Stellen Sie sicher, dass Sie your_api_token_here durch Ihr tatsächliches Wise API-Token ersetzen.
Stellen Sie außerdem sicher, dass Sie Ihre .mcp.json-Datei an den von Ihnen gewählten Modus anpassen. Wir stellen Vorlagendateien bereit, die Sie verwenden können:
Für den stdio-Modus (Standard):
cp .mcp.json.stdio .mcp.jsonFür den HTTP-Modus:
cp .mcp.json.http .mcp.jsonDiese Vorlagendateien enthalten die entsprechende Konfiguration für jeden Modus.
Der Server stellt die folgenden MCP-Ressourcen bereit:
list_recipientsGibt eine Liste aller Empfänger aus Ihrem Wise-Konto zurück.
Parameter:
profile_type: Der Profiltyp, für den Empfänger aufgelistet werden sollen. Einer aus [personal, business]. Standard: "personal"
currency: Optional. Filtert Empfänger nach Währungscode (z. B. 'EUR', 'USD')
get_recipient_requirementsRuft die Empfängeranforderungen für das Erstellen eines neuen Empfängers ab. Wenn Kontodetails bereitgestellt werden, werden diese anhand der Anforderungen validiert.
Parameter:
source_currency: Der Quellwährungscode (z. B. 'USD')
target_currency: Der Zielwährungscode (z. B. 'EUR')
source_amount: Der Betrag in der Quellwährung
profile_type: Der zu verwendende Profiltyp. Einer aus [personal, business]. Standard: "personal"
account: Optional. Die Empfängerkontodetails, die gegen die Anforderungen validiert werden sollen. Wenn nicht angegeben, werden die anfänglichen Kontonanforderungen zurückgegeben.
create_recipientErstellt einen neuen Empfänger mit den bereitgestellten Kontodetails.
Parameter:
profile_type: Der zu verwendende Profiltyp. Einer aus [personal, business]. Standard: "personal"
account: Die Empfängerkontodetails, die den Anforderungen der Wise-API entsprechen. Dies sollte beinhalten:
accountHolderName: Name des Kontoinhabers
currency: Zielwährungscode (z. B. 'EUR')
type: Kontotyp (z. B. 'iban', 'sort_code' usw.)
details: Objekt mit kontospezifischen Details (variiert je nach Währung und Land)
send_moneySendet Geld an einen Empfänger über die Wise-API.
Parameter:
profile_type: Der zu verwendende Profiltyp (personal oder business)
source_currency: Quellwährungscode (z. B. 'USD')
source_amount: Zu sendender Betrag in der Quellwährung
recipient_id: Die ID des Empfängers, an den das Geld gesendet werden soll
payment_reference: Optional. Referenznachricht für die Überweisung (Standard: "money")
source_of_funds: Optional. Herkunft der Gelder (z. B. "salary", "savings")
Die Konfiguration erfolgt über Umgebungsvariablen, die in der .env-Datei festgelegt werden können:
WISE_API_TOKEN: Ihr Wise API-Token (erforderlich)
WISE_IS_SANDBOX: Auf true setzen, um die Wise Sandbox-API zu verwenden (Standard: false)
MODE: MCP-Server-Transportmodus, entweder "http" oder "stdio" (Standard: stdio)
wise-mcp/
├── .env # Environment variables (not in git)
├── .env.example # Example environment variables
├── pyproject.toml # Project dependencies and configuration
├── README.md # This file
└── src/ # Source code
├── main.py # Entry point
└── wise_mcp/ # Main package
├── api/ # API clients
│ └── wise_client.py # Wise API client
├── resources/ # MCP resources
│ └── recipients.py # Recipients resource
└── app.py # MCP application setupUm neue Funktionen hinzuzufügen:
Fügen Sie neue API-Client-Methoden in src/wise_mcp/api/wise_client.py hinzu
Erstellen Sie neue Ressourcen in src/wise_mcp/resources/
Importieren und registrieren Sie die neuen Ressourcen in src/wise_mcp/app.py
Beiträge sind willkommen! Bitte zögern Sie nicht, einen Pull Request einzureichen.
MIT
Creates a new recipient with the provided account details.
| Name | Required | Description | Default |
|---|---|---|---|
| recipient_fullname | Yes | The name of the account holder. Required. | |
| currency | Yes | The currency code for the recipient account. Required. | |
| recipient_type | Yes | The type of recipient account. Required. | |
| profile_type | No | The type of profile to use. One of [personal, business]. Default: "personal" | personal |
| account_details | No | Additional recipient account details compliant with Wise API requirements. If provided, it will be updated with the required fields. |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description bears full burden. It only says 'creates' (indicating mutation) but doesn't disclose side effects, idempotency, error conditions, or authentication needs.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
A single sentence is concise, but it may be too brief for a tool with 5 parameters and nested objects. Every word earns its place, but more structure would improve clarity.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given 5 parameters, nested objects, and an output schema, the description is insufficient. It doesn't mention the output or provide context on how to properly construct the account_details object.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% with detailed descriptions for all 5 parameters. The description adds no additional meaning beyond stating 'with the provided account details,' which is already implied by the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states it creates a new recipient, which distinguishes it from sibling tools like list_recipients and get_recipient_requirements. However, it lacks nuance about what constitutes a 'recipient' in this context.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance on when to use this tool versus alternatives. For example, it doesn't mention prerequisites like first calling get_recipient_requirements to determine required fields.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Fetches recipient requirements for creating a new recipient. If account details are provided, validates the account details against the requirements.
| Name | Required | Description | Default |
|---|---|---|---|
| source_currency | No | Optional. The source currency code (e.g., 'USD') | |
| target_currency | No | Optional. The target currency code (e.g., 'EUR') | |
| source_amount | No | Optional. The amount in the source currency (e.g., 100.0) | |
| profile_type | No | The type of profile to use. One of [personal, business]. Default: "personal" | personal |
| account_details | No | Optional. The recipient account details to validate against requirements. If not provided, returns the initial account requirements. |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries full burden. It discloses basic behavior (fetch and validate) but omits side effects, error conditions, or safety traits. No contradiction exists.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, 24 words, front-loaded with purpose. No unnecessary words. Highly concise and well-structured.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given an output schema exists, the description covers core behavior. It could be improved by explicitly linking to 'create_recipient' as a prerequisite, but overall complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so baseline is 3. The description adds context that 'account_details' triggers validation, adding marginal value beyond the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb ('fetches') and the resource ('recipient requirements'), and distinguishes itself from siblings like 'create_recipient' and 'send_money' by focusing on fetching and validating requirements.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for checking requirements before creating a recipient, but does not explicitly state when to use this tool over alternatives like 'list_recipients' or 'send_money'.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Returns all recipients from the Wise API for the given profile type of current user. If a user has multiple profiles of that type, it will return recipients from the first profile.
| Name | Required | Description | Default |
|---|---|---|---|
| profile_type | No | The type of profile to list recipients for. one of [personal, business] | personal |
| currency | No | Optional. Filter recipients by currency code (e.g., 'EUR', 'USD') |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description must carry the burden. It discloses the 'first profile' behavior, which is a useful trait beyond the schema. However, it does not mention other behavioral aspects like idempotency, rate limits, or side effects.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences, concise and front-loaded with the core purpose. No unnecessary words or redundant information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The tool has an output schema (not shown) and low complexity with 2 parameters. The description covers the key behavioral point of first-profile limitation but lacks explanation of return structure, error cases, or pagination. Adequate but not thorough.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Input schema has 100% coverage for parameters, so baseline is 3. The description adds minimal meaning beyond the schema by mentioning 'profile type of current user' and 'first profile' but does not elaborate on currency filtering.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool returns all recipients for a given profile type, using the verb 'list' and specifying the resource 'Wise API recipients'. It distinguishes from sibling tools like 'create_recipient' which is a different action.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides a behavioral caveat about returning recipients from the first profile if multiple exist, which helps in understanding edge cases. However, it does not explicitly state when to use this tool versus alternatives or provide exclusion criteria.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Send money to a recipient using the Wise API.
| Name | Required | Description | Default |
|---|---|---|---|
| profile_type | Yes | The type of profile to use (personal or business) | |
| source_currency | Yes | Source currency code (e.g., 'USD') | |
| source_amount | Yes | Amount in source currency to send | |
| recipient_id | Yes | The ID of the recipient to send money to | |
| payment_reference | No | Optional. Reference message for the transfer (defaults to "money") | |
| source_of_funds | No | Optional. Source of the funds (e.g., "salary", "savings") |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries full burden, but it provides no behavioral details beyond the name. It does not disclose mutation implications, error conditions, or authorization needs.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Single sentence, front-loaded, no unnecessary words. Every word earns its place.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool has 6 parameters and no annotations, the description is too brief. It lacks context about success/failure, idempotency, and usage flow. Output schema exists but does not compensate for missing behavioral context.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so baseline is 3. The description adds no extra context about parameter meaning beyond what the schema already provides.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb 'Send' and resource 'money to a recipient using the Wise API', distinguishing it from sibling tools that manage recipients (create, get requirements, list).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is provided on when to use this tool versus alternatives. It does not mention prerequisites, such as needing a recipient created first, or when not to use it.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.0create_recipientget_recipient_requirementslist_recipientssend_moneyScored across 4 tools
Each tool has a clearly distinct purpose: creating recipients, fetching requirements, listing recipients, and sending money. No overlap.
All tool names follow a consistent verb_noun pattern in snake_case (e.g., create_recipient, list_recipients, get_recipient_requirements, send_money).
With 4 tools, the server is well-scoped for a focused Wise integration covering recipient management and money transfer.
Covers key recipient and transfer operations, but lacks a direct get_recipient by ID or update/delete recipient, which are minor gaps.
MCP server for Modern Treasury — payment orders, transactions, counterparties and ledgers.
The Mercado Pago MCP Server implements the Model Context Protocol to provide AI agents and LLMs with access to Mercado Pago's APIs and tools within compatible development environments. It acts as an intermediary that translates Mercado Pago resources into executable functions (tools) that AI applications can invoke to perform actions and automate flows. The server simplifies integration, enables using documentation to implement or improve code, and optimizes operations through natural language interactions without manual implementations.
A paid remote MCP for AI SDK MCP gateway registry, built to return verdicts, receipts, usage logs, a
Zero-setup MCP gateway securely connecting AI to your tools with authentication and workflows
