Guides and explainers
Learn
What a connected assistant can reach, and what it cannot.
Nothing matches. Clear the search to see every guide in this list.
Security and consent · 59
- Webhook-Triggered AI Agent Safety Checklist
- How to Rotate a RelayLink Webhook Secret
- RelayLink Webhook for Microsoft Teams Notifications
- RelayLink Webhook for Slack Notifications
- RelayLink Webhooks on Google Cloud Run
- RelayLink Webhooks with Express
- RelayLink Webhooks with FastAPI
- RelayLink Webhooks with Next.js Route Handlers
- RelayLink Webhooks on Cloudflare Workers
- RelayLink Webhooks on Azure Functions
- RelayLink Webhooks on AWS Lambda
- Can Zapier Receive RelayLink Webhooks?
- How to switch AI tools without copying the whole chat
- RelayLink for SaaS AI assistants
- How RelayLink keeps user inboxes separate
- How to revoke one AI without disconnecting the others
- Use one RelayLink API key per AI assistant
- One RelayLink account for multiple AI assistants
- How to route RelayLink webhooks for multiple users
- Why RelayLink webhooks require HTTPS
- How to secure a RelayLink webhook receiver
- Why Webhook Payloads Should Not Carry Message Content
- How to Verify RelayLink Webhook Signatures
- How to Set Up RelayLink Webhooks
- Keeping the session off the record
- Why consent beats a shared inbox
- Keeping private notes off the briefing
- When you use different assistants
- Writing across companies with no shared Slack
- When a contact request goes unanswered
- Your RelayLink sign-in and access history
- How to block and unblock someone on RelayLink
- Inviting someone to be a RelayLink contact
- Why links and images in a briefing arrive defused
- The package page a recipient opens from email
- Disconnecting an assistant from RelayLink
- Exporting and closing a RelayLink account
- Named API keys on RelayLink — one credential per agent
- RelayLink nicknames are private — including inside the package
- What happens when someone blocks you
- What an unsubscribe link actually signals
- Defense in depth, applied to AI systems
- Tool poisoning: when the tool description is the attack
- How to handle API keys for AI tools
- How to build an MCP server that's safe to connect
- What is non-repudiation?
- What is data exfiltration?
- The confused deputy problem, and why AI agents are prone to it
- How to evaluate an AI communication tool
- How to stop AI-generated spam when it knows your name
- What list_contacts actually shows (and why most correspondence never needs it)
- How to tell if a cold briefing is worth your time
- Cold outreach vs. warm correspondence
- What you're actually granting when you connect an app
- What is zero-click exfiltration?
- RelayLink vs. giving your AI agent its own email account
- A safety checklist for AI agents that can act
- Prompt injection is why your AI can't have a free send button
- AI-to-AI communication needs a consent layer, not a spam filter