Engineering notes on the self-improving firewall.

Your AI agent is a supply chain attack surface. Here's how to gate it.

Your AI coding agent runs shell commands, installs packages, modifies files, and calls external APIs. Static scanners audit output after the fact. ThumbGate runs pre-action checks via PreToolUse hooks so the action is inspected before execution. Thumbs-down feedback feeds a promotion path: repeated patterns can become warning or blocking gates.

Start with npx thumbgate init. Gates are reviewable local policy—commit them, share them, tighten them from real incidents.

Dual-signal feedback: why “what failed” isn’t enough

Optional failureType on capture separates wrong decision from bad execution so reliability signals can diverge by dimension. Backward compatible with plain thumbs-down.

The Claude Code leak proves why pre-action checks matter

Permissions, hooks, retry loops, and local memory are product surfaces. Treat them as governed workflow boundaries. ThumbGate adds feedback-driven pre-action checks so repeated mistakes can be flagged—and catastrophic classes hard-blocked—before the next tool call.

v0.8.5: gate reasoning chains and funnel fixes

Blocks and warnings explain which pattern matched, the gate identity, and whether the source was manual policy or auto-promotion. CLI-first users get upgrade paths without relying only on the website checkout page.