Stop switching tabs to stay secure.
Bring Aribot’s AI security agents into ChatGPT, Gemini, Claude and your CLI. Just ask — threat models, code & API security, shadow-AI, cloud & compliance, fixes — right where you already work.
How it works
Connect once
Add Aribot in your assistant’s connector settings and approve it once. It’s a secure sign-in — you never paste keys or tokens.
Just ask
Ask in plain language — “threat-model this design”, “scan this repo”, “find shadow AI”. Answers come back in the chat and in the Aribot app.
Private & governed
The agent only ever acts inside your organisation. Your admin decides who can connect and which agents are on — no data crosses between companies.
Top up in the flow
Reach a limit? You get a secure link right there to add credit or request a plan — from the app, the CLI, or the assistant.
Connect — step by step
- Open Settings → Connectors.
- Click Add custom connector.
- Paste the server URL:
https://mcp.aribot.ayurak.com/ - Sign in to Aribot and approve consent (a company Owner or Security Manager approves the link, once).
- The Aribot tools now appear in Claude — start asking.
- Open Settings → Connectors.
- Click Add and paste the same MCP URL.
- Approve the one-time secure sign-in.
- Prefer a custom GPT? Import the OpenAPI schema:
https://api.aribot.ayurak.com/.well-known/openapi.json
- Register the OpenAPI schema as an OAuth 2.0 tool:
https://api.aribot.ayurak.com/.well-known/openapi.json - Or add the remote MCP server URL above.
- Approve the secure sign-in to link your Aribot workspace.
- Add the remote MCP server URL:
https://mcp.aribot.ayurak.com/ - Sign in once when prompted — the tools load automatically.
Server (MCP): https://mcp.aribot.ayurak.com/ · OpenAPI: https://api.aribot.ayurak.com/.well-known/openapi.json
Secure by default: OAuth 2.0 + PKCE, scoped to your company. Only an Owner or Security Manager can approve the link.
What you can ask it to do
Turn an architecture diagram or a description into a full STRIDE threat model — prioritized threats, security requirements and control mappings.
Scan an upload for vulnerabilities and verify whether the modeled threats are actually mitigated in your source code.
Inventory every API endpoint with its authentication status, risk level and risk factors — and assign ownership.
Discover unsanctioned or unknown AI and API usage across your code before it becomes an exposure.
Posture checks across AWS, Azure and Google Cloud, aligned to CIS and NIST, per account.
Report control and framework coverage — NIST, ISO 27001, SOC 2 — and your live compliance status.
Follow the full chain: threat → requirement → control → remediation, end to end.
Get prioritized, context-aware fix guidance. Preview a fix for free; you decide whether to apply it.
Viewing results is always free — you only pay when an agent runs an action. Every organisation’s data stays private: the agent only ever acts within the organisation you belong to.