Spurlock Studios
Contact
Tag

security

Posts tagged security.

Sandboxed Tool Use: Letting Agents Act Without Letting Them Loose AI Agents

Sandboxed Tool Use: Letting Agents Act Without Letting Them Loose

Tool use without a sandbox is an API key with opinions. Allowlists, scoped credentials, blast-radius caps, and human gates are how agents earn the right to touch production.

10 MIN
OAuth Tokens Will Expire: Stop Silent 401 Loops in Production Automation

OAuth Tokens Will Expire: Stop Silent 401 Loops in Production

n8n refreshes OAuth reactively on tokenExpiredStatusCode (default 401). Pause on auth failure, prefer service accounts, and stop quiet overnight 401 loops.

10 MIN
Prompt Injection for Tool Agents: Stop Text from Becoming Actions AI Agents

Prompt Injection for Tool Agents: Stop Text from Becoming Actions

Defend production agents that read email, tickets, or pages by fencing untrusted text as data and blocking injection paths into tool calls.

12 MIN
Webhook Security for Automations: Signatures, Secrets, and Least Privilege Automation

Webhook Security for Automations: Signatures, Secrets, and Least Privilege

Secure n8n webhooks with signature verification, secret handling, least-privilege credentials, and replay defenses that belong in production.

10 MIN
Pre-Execution Policy Gates: The Kill Switch That Lives Outside the Prompt AI Agents

Pre-Execution Policy Gates: The Kill Switch That Lives Outside the Prompt

Your agent’s kill switch is a pre-execution policy gate outside the prompt: allow, deny, or pending-approval before side effects — fail closed on outages.

11 MIN