Stop paying for the same AI mistake twice.
Get Started with a $499 Diagnostic (one hard gate + proof) or self-serve Pro at $19/mo. Free local evaluate stays free. Countable units — not vibes.
npx thumbgate init — first hard gate usually minutes after install.Make the alternative look expensive.
Same move as high-converting SaaS pricing: DIY prompts cost nothing and stop nothing; platform hire and GRC suites cost months.
| DIY / CLAUDE.md | Platform hire | GRC suite | ThumbGate | |
|---|---|---|---|---|
| Monthly cost | $0 | $15k+ loaded | $$$$ | $0–$19 · $499 once |
| Time to first gate | Never reliable | Weeks | Months | Minutes after install / 2 business days managed |
| When agent “looks competent” | Ignores the prompt | Custom glue | Gateway only | Hard block at tool call |
A bounded implementation, not a strategy deck.
The goal is one working enforcement boundary with a test that proves the original mistake cannot silently recur.
Review the failure
Reproduce the risky action and identify the exact point where an external check can intervene.
Install the gate
Configure ALLOW, WARN, or DENY behavior for the supported local agent workflow.
Prove the boundary
Add a regression test and hand back rollout and rollback evidence.
Know the boundary before you pay.
Good fit
A repeated, concrete agent action with a known command, tool call, destination, repository, or resource boundary.
Enterprise boundary
The $499 offer covers one workflow. Org-wide hosted policy, SSO, a hosted audit dashboard, 24/7 incident response, and multi-workflow rollout require separate scope. Hosted Enterprise capabilities are not generally available.