A writeup of a small agentic system I built to triage my Gmail inbox, using Claude subagents to organize and assess messages.
#security
7 posts tagged “security”. Follow along via the #security RSS feed.
Triaging Gmail with Claude Subagents You Need to Rewrite Your CLI for AI Agents Human DX optimizes for discoverability. Agent DX optimizes for predictability. What I learned building a CLI for agents first.
Abandon Git LFS because AI Agents Git LFS causes fatal errors in AI agents like Jules due to proxy conflicts and hook limitations. Learn why I abandoned LFS and migrated back to standard Git.
Currentonly Scopes in Google Apps Script How Apps Script currentonly scopes limit access to an active file, and why Advanced Services, direct API calls, and openById need full scopes.
Secure Secrets in Google Apps Script Avoid hardcoded Apps Script secrets by using PropertiesService for configuration and Google Cloud Secret Manager for sensitive values.
Protect Gmail AI Agents with Model Armor Use Model Armor in an MCP tool handler to block prompt injection and redact sensitive Gmail content before it reaches an AI agent.
Apps Script Service Account Impersonation Avoid downloading service-account keys: use impersonation in Apps Script to mint access tokens for Google Cloud APIs.