MCP for developers
SanctionsKit MCP server for AI coding agents
Connect SanctionsKit to your coding agent so it can read the docs, write your integration, and test it with sample data.
Build your integration with AI
A ready-to-use prompt gives your AI the docs, setup steps, and testing plan. Choose where you want to start.
“Help me integrate SanctionsKit into my application.”
Desktop agents work in your project. Google Gemini opens in AI Studio for browser-based help. Review the prompt, then send it when you’re ready.
View or copy the integration prompt
Help me integrate SanctionsKit into my application.
I want sanctions screening with source and match evidence that my team
can review, using SanctionsKit's documentation and sandbox tools to
build and test the integration in my project's existing stack.
Start here:
- MCP setup: https://www.sanctionskit.com/guides/mcp-integration
- API quickstart: https://www.sanctionskit.com/docs/quickstart
- API schema: https://www.sanctionskit.com/openapi.json
- MCP server (Streamable HTTP): https://www.sanctionskit.com/mcp
If no project is open, ask me to select one. Inspect existing MCP
configuration before changing it. If SanctionsKit is not connected,
help me add it using this client's supported setup without replacing
other connections. Use the guide's Codex configuration only in Codex;
use the current client's documentation for other clients. Let me
complete any required sign-in or securely configure a sandbox key.
Never ask me to paste a credential into this conversation.
Use the SanctionsKit MCP to add server-side screening to this app.
1. Read the quickstart, authentication, screening, errors, and
idempotency docs. Get the current API schema before writing code.
2. Inspect this project's stack and follow its existing conventions.
3. Check my sandbox profile and usage, and inspect source metadata.
4. Validate an invented person: Alex Morgan, birth year 1984,
with package sandbox@1. Confirm synthetic:true in the sandbox
screening tool call. Run one sandbox screening and retrieve it.
Create one idempotency key for this operation and reuse that key
with the identical request if a network retry is needed.
5. Build a server-side REST integration. Read its API key from secure
environment configuration; never print the key or subject inputs.
6. Handle potential_match, no_match, and request failures separately.
A no-match is limited to selected coverage, not legal clearance.
7. Test success, unavailable coverage, invalid credentials, rate
limits, timeouts, and idempotent retries with synthetic fixtures.
8. Explain the files changed, tests run, and remaining setup.
Use only synthetic subjects. Do not run production screenings,
change account settings, or deploy the application.
If MCP, authentication, or the sandbox is unavailable, continue with
the public docs, schema, and local synthetic tests where possible.
Report the blocked connection or sandbox check explicitly; do not
invent tool calls, test results, or source availability. Keep MCP
credentials separate from the application's server-side REST key.On this page
Quick start
MCP (Model Context Protocol) lets your coding agent use a service while it works on your code. Connect SanctionsKit, then ask the agent to add screening to your app. It can look up the API, write the integration, and test it with invented data.
Use the server URL below. The step-by-step guide gives you the exact Codex configuration and sign-in command.
https://www.sanctionskit.com/mcp- 1. Add SanctionsKit to your coding agent using the guide.
- 2. Sign in to your SanctionsKit workspace to enable sandbox tests. Documentation is available without signing in.
- 3. Open your project and paste the prompt below.
Try it in your project
Start with this prompt. Add the feature you want, such as checking a supplier during onboarding. Your agent should explain the files it changes and the tests it runs.
Use the SanctionsKit MCP to add screening to this app.
Read the current docs and follow this project's conventions.
Build a server-side integration and test it with invented sandbox data.
Keep credentials out of the code, conversation, and logs.
Show me the changes and test results. Do not deploy.What your agent can do
The MCP brings SanctionsKit documentation and development tools into the conversation. You keep working in your own editor and reviewing your own code.
- Find the right documentation and check current request and response fields.
- Validate a request before sending it.
- Run an invented screening in your sandbox, read the result, and check your remaining allowance.
- Use what it learns to build and test your application’s REST API integration.
Sign in or use an API key
For everyday use, sign in to SanctionsKit when connecting your agent and approve its sandbox access. This connection method is called OAuth. You do not need to copy an API key.
For a compatible direct client or an automated job, you can supply a sandbox API key from your environment or secret store instead. Both methods connect to your SanctionsKit workspace. The developer guide explains both options after the quick start.
From a sandbox test to your application
The sandbox is a test environment with invented records. A screening uses sandbox allowance and saves evidence under your retention policy; it may also create a review case. MCP account tools are sandbox-only. Keep real customer inputs out of these development tools.
Your application uses its own server-side SanctionsKit API key once the integration is ready. Review the generated code and tests, then choose your production coverage through your normal release process. A potential match needs human review; a no-match is limited to the selected coverage and is not legal clearance.