--- title: "Connect Sorry to Lakera AI Guardrails - APIANT.ai" description: "When \"New notice\" fires in Sorry, your agent runs \"Check for Jailbreak Attempt\" in Lakera AI Guardrails." canonical: "https://apiant.ai/connect/sorry-to-lakera-ai-guardrails" --- # Connect Sorry to Lakera AI Guardrails When **New notice** fires in [Sorry](https://apiant.ai/connections/sorry), your agent runs **Check for Jailbreak Attempt** in [Lakera AI Guardrails](https://apiant.ai/connections/lakera-ai-guardrails). Describe the whole flow and it is built, tested on every scenario and kept running. Sorry **1** trigger Sorry **16** actions Lakera AI Guardrails **5** actions What the catalog holds today. Ask for anything missing and your agent reads the API documentation and adds it. ## What do you want Sorry and Lakera AI Guardrails to do together? Give your AI client this address. `https://mcp.apiant.ai` One connection does both: ask your client to act in an app once, or have it build an automation that keeps running. Any client that takes a remote MCP server. Nothing to install. Create your free account, then sign in once in your browser. ### Set it up in your client ### Claude Code Run this in the directory you want to work in. `claude mcp add --transport http -s project apiant-ai https://mcp.apiant.ai` Start Claude Code there, approve the server, and sign in in the browser page it opens. [Claude Code guide (opens in a new tab)](https://apiant.ai/docs/mcp/claude-code) ### Claude, web and desktop Open Customize, then Connectors, then Add custom connector, and paste the address. On Team and Enterprise an owner adds it first. [Setup guide (opens in a new tab)](https://apiant.ai/docs/mcp) ### Cursor Add the address to your MCP configuration as a server URL. Cursor signs you in in your browser. [Setup guide (opens in a new tab)](https://apiant.ai/docs/mcp) ### Zed Open Settings, then AI, then MCP Servers, then Add Server, then Add Remote Server. [Setup guide (opens in a new tab)](https://apiant.ai/docs/mcp) ### Codex and the ChatGPT desktop app Open Settings, then MCP servers, then Add server, and choose Streamable HTTP. From the terminal, Codex takes the address with its own add command and then signs you in. [Setup guide (opens in a new tab)](https://apiant.ai/docs/mcp) ### Any other MCP client Paste the address where the client asks for a remote MCP server URL, and name the server apiant-ai where it lets you. APIANT's skills reach the client over the same connection. [Setup guide (opens in a new tab)](https://apiant.ai/docs/mcp) Example flows ## What Sorry can start in Lakera AI Guardrails - When **New notice** in Sorry then **Check for Jailbreak Attempt** in Lakera AI Guardrails - When **New notice** in Sorry then **Check for Prompt Injection** in Lakera AI Guardrails Starting points drawn from the triggers and actions in the catalog today. Describe the flow you need, at any depth, and your agent builds it and tests every scenario before it ships. Everything your agent can do with Sorry and Lakera AI Guardrails ## Triggers and actions 1 trigger · 21 actions - Trigger **New notice** Triggers when a new notice (incident, maintenance or general announcement) is created on the selected Sorry status page. - Action **Add component** Creates a new component on a Sorry status page, optionally nested under a parent component. - Action **Add notice** Creates a new incident, maintenance or general notice on a Sorry status page, optionally linking affected components and publishing it to subscribers. - Action **Add notice update** Posts a new update on an existing notice on a Sorry status page, optionally publishing it to your configured integrations. - Action **Add subscriber** Adds an email or SMS subscriber to a status page, optionally limited to specific components, so they receive status notices. - Action **Delete notice** Deletes a notice (incident, maintenance or general) from a Sorry status page. - Action **Delete subscriber** Permanently removes a subscriber from a status page. - Action **Get notice** Returns a single notice from a status page, including its type, state, subject, synopsis and link. - Action **Get status page** Returns a single Sorry status page with its name, state, timezone, support details and notification settings, optionally including related brand, components, notices and more. - Action **List components** Returns all components (the services or infrastructure elements) shown on a Sorry status page, with their current state. - Action **List notices** Returns the notices (incidents, maintenance and general notices) posted on a Sorry status page, optionally filtered by type, state, subject or creation date. - Action **List notice updates** Returns all updates posted on a notice on one of your Sorry status pages, optionally filtered by creation date or content. - Action **List status pages** Returns all status pages in your Sorry account, optionally filtered by name or creation date. - Action **List subscribers** Returns the subscribers on a status page, optionally only those who receive notices for a specific component. - Action **Update component** Updates a component on a Sorry status page, such as its name, description, position or parent component. - Action **Update notice** Updates an existing notice on a Sorry status page — change its subject or state, add a comment, set affected components or reschedule planned maintenance. - Action **Update status page** Updates a status page's settings, such as its name, timezone, support contact details, search visibility and language. - Action **Check for Jailbreak Attempt** Screens a piece of text with Lakera AI Guardrails and returns whether it was flagged as a jailbreak or prompt attack, with the per-detector breakdown. - Action **Check for Prompt Injection** Screens a piece of text with Lakera AI Guardrails and returns whether it was flagged, with a per-detector breakdown so prompt injection detections can be checked. - Action **Detect PII** Screens text with AI Guardrails and returns any personal information (PII) found, with its type and location in the text. - Action **Detect Toxicity** Screens text for toxic, hateful, violent, sexual or profane content using your AI Guardrails project policy and returns whether it was flagged. - Action **Screen Content** Screens an LLM conversation (and optionally tool definitions) against your AI Guardrails project policy and returns whether it was flagged, with optional per-detector breakdown and match locations. ## Each app on its own - [ **Sorry** Status page and incident communication platform for notifying customers about downtime.](https://apiant.ai/connections/sorry) - [ **Lakera AI Guardrails** Lakera AI Guardrails is a security platform detecting prompt injections, jailbreaks, PII, toxicity, and...](https://apiant.ai/connections/lakera-ai-guardrails) Reviews ## What our customers say ## More from Sorry - [ **Sorry to Stripe** Stripe is a payment processing platform for online businesses.](https://apiant.ai/connect/sorry-to-stripe) - [ **Sorry to Trello** Trello is a visual collaboration tool that organizes projects into boards, lists, and cards.](https://apiant.ai/connect/sorry-to-trello) - [ **Sorry to Zoom** Zoom unifies cloud video conferencing, simple online meetings, group messaging, and a software-based...](https://apiant.ai/connect/sorry-to-zoom) - [ **Sorry to Twilio** Cloud communications platform for sending SMS/MMS messages, making and managing voice calls, and looking up...](https://apiant.ai/connect/sorry-to-twilio) - [ **Sorry to Typeform** Build and publish online forms, surveys, and quizzes with Typeform.](https://apiant.ai/connect/sorry-to-typeform) - [ **Sorry to Webflow** Webflow is a professional website builder and CMS for designing, building, and launching responsive...](https://apiant.ai/connect/sorry-to-webflow) ## More into Lakera AI Guardrails - [ **HubSpot to Lakera AI Guardrails** HubSpot is your all-in-one stop for all of your marketing software needs.](https://apiant.ai/connect/hubspot-to-lakera-ai-guardrails) - [ **Salesforce to Lakera AI Guardrails** Salesforce is a leading enterprise customer relationship management (CRM) platform.](https://apiant.ai/connect/salesforce-to-lakera-ai-guardrails) - [ **Slack to Lakera AI Guardrails** Slack is a messaging platform for teams. Post messages, manage channels, users, reminders, and react to...](https://apiant.ai/connect/slack-to-lakera-ai-guardrails) - [ **Google Sheets to Lakera AI Guardrails** Google Sheets](https://apiant.ai/connect/google-sheets-to-lakera-ai-guardrails) - [ **Gmail to Lakera AI Guardrails** Gmail is Google’s email service, featuring spam protection, search functions, and seamless integration with...](https://apiant.ai/connect/gmail-to-lakera-ai-guardrails) - [ **Google Calendar to Lakera AI Guardrails** Google Calendar is a time management tool providing scheduling features, event reminders, and integration...](https://apiant.ai/connect/google-calendar-to-lakera-ai-guardrails) Questions ## Sorry and Lakera AI Guardrails, answered ### ### What can start an automation from Sorry? Any of Sorry's 1 trigger in the catalog, including New notice. Describe the moment in plain language and your agent picks the right one. ### ### What can your agent do in Lakera AI Guardrails when something happens in Sorry? Any of Lakera AI Guardrails's 5 actions, including Check for Jailbreak Attempt, Check for Prompt Injection, Detect PII and Detect Toxicity. Your agent maps the data both ways and tests every scenario before it ships. ### ### Can it also run from Lakera AI Guardrails to Sorry? Lakera AI Guardrails has no triggers in the catalog today, so automations start in Sorry. If you need the other direction, ask: your agent reads Lakera AI Guardrails's API documentation and adds it. ### ### Do I have to build a connector for Sorry or Lakera AI Guardrails? No. Both are in the catalog with their triggers and actions. You describe the automation; your agent builds it, tests every scenario and keeps it running. ### ### Can my AI client use Sorry and Lakera AI Guardrails without building an automation? Yes. With your AI client connected to APIANT, ask it to do something in either app once and it runs that action on the spot. Ask for an automation and your agent builds it, tests every scenario and keeps it running. One connection covers both. ### ### What happens when something fails? Your agent works that out for you. It plans the failure paths along with the happy path, then tests every scenario before it ships. ### ### Is there a free plan? Yes. Every plan includes the agent, including free, and building from your own AI client spends no assistant credits.