AI Agent Guardrails Bundle

Guardrails for what agents DO, not just what they say: block risky tool calls before they run, make the ones that run recoverable, and raise autonomy only when the record earns it.

$69 one-time

Buy and download

Pay with crypto (USDC or USDT)

Same price on Base, Ethereum, Arbitrum, Optimism, Polygon or Solana. No account, no KYC. Files appear as soon as the payment confirms. How it works

14-day refund: if it doesn’t work and we can’t fix it, email hello@youraitools.co within 14 days and we refund you.

Sold by YourAITools. Support: hello@youraitools.co

Delivery: Right after payment, Stripe’s confirmation page shows your download link.
Format and license: Zip, 87 KB, v1.0, 67 files. Each kit keeps its own license: use and modify the kits for your own work and for clients you serve; do not resell, sublicense or publish the kits themselves.
You need: Python 3.9+ for the Python kits; Node 20+ for the TypeScript kits. Tests run offline except for package installs.
Refunds: 14-day refund: if it doesn’t work and we can’t fix it, email hello@youraitools.co within 14 days and we refund you. Refund policy

Download$39

Claude Agent SDK Guardrails Kit

Tool policy, single-use approvals, spend caps and offline hook tests for the Claude Agent SDK (Python).

View Claude Agent SDK Guardrails Kit

Download$59

LangGraph Recoverable Actions Starter

Human-approved LangGraph tool calls that survive a timeout or crash without repeating the write.

View LangGraph Recoverable Actions Starter

Download$19

Agent Safety Skills

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

View Agent Safety Skills

Download$7

Autonomy Ladder

Human-in-the-loop policy: per action, your agent acts, asks or never acts.

View Autonomy Ladder

Who it’s for

Built for teams shipping Claude Agent SDK or LangGraph agents that touch real systems.

What’s inside

Bought separately these cost $124. Each kit is the exact, unchanged version sold on its own page.

Claude Agent SDK Guardrails Kit: Tool policy, approvals, spend caps and offline tests for the Claude Agent SDK (Python) ($39 alone)
LangGraph Recoverable Actions Starter: LangGraph human-approved tool calls that survive crashes and retries ($59 alone)
Agent Safety Skills: 4 skills + scripts: prompt-injection, spend and secrets guardrails ($19 alone)
Autonomy Ladder: Human-in-the-loop policy: what an agent may do alone, with approval, or never ($7 alone)

What you get

One zip with every kit in components/, run_all_tests.sh (runs each kit’s own tests in place) and MANIFEST.sha256 (the sha256 of each kit’s original download).

Real run_all_tests.sh log for the AI Agent Guardrails Bundle

How they fit

Hooks (block) -> recoverable actions (undo) -> safety skills (secrets, approvals) -> autonomy ladder (when to loosen).

Quick start

Open the README of the kit you need first (links above). Each one runs on its own.
./run_all_tests.sh runs every kit's own test suite in place and prints one line per kit. Kits that need packages (npm, pip) install them into their own folder (node_modules/, .venv/); nothing is installed globally.
MANIFEST.sha256 lists the sha256 of each kit's original download, so you can check it against a fresh copy.

Honest limits

Each kit’s own README ‘Honest limits’ apply unchanged.
The bundle is the same files as the individual products (no extra code), so a buyer who owns one kit is paying for the rest.
Updates ship per kit.

Requirements

Python 3.9+ for the Python kits; Node 20+ for the TypeScript kits. Tests run offline except for package installs.

Format and license

Zip, 87 KB, v1.0, 67 files. Each kit keeps its own license: use and modify the kits for your own work and for clients you serve; do not resell, sublicense or publish the kits themselves.

FAQ

Does it replace NeMo Guardrails / Guardrails AI?: No; those check text. Use these for tool calls and actions.
Claude Agent SDK only?: The SDK kit is Claude-specific; the LangGraph kit is LangGraph; the two skill packs work with any agent that reads SKILL.md.
Is it a security guarantee?: No; it's tested patterns that reduce risk.
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.

Get AI Agent Guardrails Bundle

AI agent guardrails you can test: Claude Agent SDK hooks, LangGraph undo/approval patterns, agent safety skills and an autonomy ladder. 4 kits, $69 ($124 separately).

$69 one-time

Buy and download Pay with crypto

Pairs well with

Download$59

x402 Developer Bundle

Build with x402 payments: builder skills, a paid agent API starter, Hermes and ElizaOS plugin templates. 4 tested kits, $59 ($96 separately).

View x402 Developer Bundle

Download$79

AI Automation Agency Toolkit

Tools to deliver AI automation for clients: AI receptionist booking backend (Retell + Vapi), n8n agent workflows, change control and RFP evidence skills. $79 ($116 separately).

View AI Automation Agency Toolkit