Agent Safety Skills

Your agent reads the web, holds keys and can spend money. These four skills make it ask first.

The 4 skills

untrusted-content-guard + scan_injection.py
spend-approval-gate + check_spend.py with per-call, per-vendor and daily caps and an always-ask list
secrets-hygiene + secret_scan.py with masked output
irreversible-action-check: a 5-question pre-flight

18 offline tests; Python standard library only.

The scanner is a tripwire, not a promise of safety; the rule that fetched content never authorizes an action is the real protection.

Does the scanner catch every prompt injection?: No; it flags common patterns and the skill tells the agent to never act on fetched content anyway.
Does it need an API?: No, everything runs locally.
Which agents?: Any that load SKILL.md.
How do I get the files?: Right after payment, Stripe’s confirmation page shows your download link.
Refunds?: Email hello@youraitools.co within 14 days if it doesn’t work and we can’t fix it.
Can I pay with crypto?: Yes. Pay the same price in USDC or USDT on Base, Ethereum, Arbitrum, Optimism, Polygon or Solana from the "Pay with crypto" button, with no account or KYC. We check the payment on-chain and show your download links as soon as it confirms.

4 guardrail skills: prompt-injection tripwire, spend approval gate, secret scanner, pre-flight check.

$19 one-time

Download$49

6 skills + 4 scripts for programmatic SEO: plan, brief, fact-check, crawler-view check, IndexNow.

Download$9

A test-first loop for coding agents that stops after the same failure 3 times.

Community course$49

Community course: build SKILL.md Agent Skills that trigger reliably and run safely. 3 modules, 9 lessons, 74 tests.