# Cryptographic Guardrails for AI Agents | ICME Labs ## Home - [Home](https://docs.icme.io/home.md): Welcome to your team’s developer platform ## Documentation - [ICME PreFlight API](https://docs.icme.io/documentation/docs.md): Cryptographic AI agent guardrails with automated reasoning, formal verification, and zero-knowledge proofs for verifiable agentic commerce. - [Our Products](https://docs.icme.io/documentation/our-products.md) - [Verbatim Deterministic Gateway](https://docs.icme.io/documentation/our-products/verbatim-deterministic-gateway.md): Your AI agent can't hallucinate. Not "rarely." Can't. - [Prompt Injection Instruments](https://docs.icme.io/documentation/our-products/prompt-injection-instruments.md) - [Vericoding](https://docs.icme.io/documentation/our-products/vericoding.md): Smart contracts in English, verified by math. Type what your contract should do. Get verified code that provably does it. Deploy. Share the proof. - [Walkthrough 1: Freelancer Payment](https://docs.icme.io/documentation/our-products/vericoding/walkthrough-1-freelancer-payment.md): An end-to-end example of vericoding with ICME PreFlight and Certora. - [Quickstart](https://docs.icme.io/documentation/getting-started/quickstart.md): Get your first policy compiled and your first action checked in under 5 minutes. - [Pricing](https://docs.icme.io/documentation/getting-started/pricing.md): ICME uses pay-per-use pricing. There are no subscriptions, no seats, no monthly minimums. You pay for what you use, when you use it. - [Stop Watching Your Agents. Start Proving They're Right.](https://docs.icme.io/documentation/getting-started/stop-watching-your-agents.-start-proving-theyre-right..md): How cryptographic proofs strengthen agent observability by making every decision provably correct, tamper-proof, and privately auditable. - [How ICME PreFlight Works](https://docs.icme.io/documentation/learning/how-icme-preflight-works.md): Learn how ICME PreFlight uses automated reasoning, formal verification, and zero-knowledge proofs to secure AI agents against prompt injection and unsafe actions. - [What ICME PreFlight defends against](https://docs.icme.io/documentation/learning/how-icme-preflight-works/what-icme-preflight-defends-against.md) - [Relevance Screening](https://docs.icme.io/documentation/learning/how-icme-preflight-works/relevance-screening.md) - [Formal Verification vs Prompt-Based Guardrails](https://docs.icme.io/documentation/learning/how-icme-preflight-works/formal-verification-vs-prompt-based-guardrails.md) - [Zero-Knowledge Proofs](https://docs.icme.io/documentation/learning/how-icme-preflight-works/zero-knowledge-proofs.md) - [Understanding SAT / UNSAT](https://docs.icme.io/documentation/learning/how-icme-preflight-works/understanding-sat-unsat.md) - [Battle Testing](https://docs.icme.io/documentation/learning/battle-testing.md): Tools to help make your rules cover 100%. - [Writing Effective Policies](https://docs.icme.io/documentation/learning/battle-testing/writing-effective-policies.md) - [JOLT Atlas (zkML)](https://docs.icme.io/documentation/learning/jolt-atlas-zkml.md) - [ONNX](https://docs.icme.io/documentation/learning/jolt-atlas-zkml/onnx.md) - [ONNX Runtime](https://docs.icme.io/documentation/learning/jolt-atlas-zkml/onnx/onnx-runtime.md) - [Quantization](https://docs.icme.io/documentation/learning/jolt-atlas-zkml/onnx/quantization.md) - [Tensor Instructions](https://docs.icme.io/documentation/learning/jolt-atlas-zkml/onnx/tensor-instructions.md) - [Lookups](https://docs.icme.io/documentation/learning/jolt-atlas-zkml/onnx/lookups.md) - [Architecture](https://docs.icme.io/documentation/learning/jolt-atlas-zkml/architecture.md) - [AGENTIC COMMERCE](https://docs.icme.io/documentation/use-cases/agentic-commerce.md) - [Crypto Wallet Agent Protection](https://docs.icme.io/documentation/use-cases/agentic-commerce/crypto-wallet-agent-protection.md): AI agents with signing authority over wallets are losing real money to attacks that have nothing to do with breaking cryptography. - [Protocol, Treasury, and Signer Protection](https://docs.icme.io/documentation/use-cases/agentic-commerce/protocol-treasury-and-signer-protection.md): Cryptographic policy enforcement for protocols, DAOs, bridges, custodians, and institutional treasuries. Uses natural-language rules, proofs for every action, and receipts any counterparty can verify. - [Verifiable API Discovery: How Buyer Agents Find the Right Tool](https://docs.icme.io/documentation/use-cases/agentic-commerce/verifiable-api-discovery-how-buyer-agents-find-the-right-tool.md) - [Cryptographic Guardrails for Claude Code](https://docs.icme.io/documentation/use-cases/agentic-commerce/verifiable-api-discovery-how-buyer-agents-find-the-right-tool/cryptographic-guardrails-for-claude-code.md): A secure alternative to hookify for Claude Code. - [Agentic Commerce Clearing House](https://docs.icme.io/documentation/use-cases/agentic-commerce/agentic-commerce-clearing-house.md): A single settlement policy that verifies both parties' guardrail proofs before any x402 stablecoin transfer settles. No trust required. - [Mastercard Verifiable Intent](https://docs.icme.io/documentation/use-cases/agentic-commerce/mastercard-verifiable-intent.md): Integrate Preflight with Mastercard's Verifiable Intent, the cryptographic trust layer in Mastercard Agent Pay. Embed formally verified enterprise-policy compliance proofs in VI Layer 3 credentials. - [Circle Nanopayments](https://docs.icme.io/documentation/use-cases/agentic-commerce/circle-nanopayments.md): Spending policy proofs for AI agents transacting in USDC. - [PRIVACY & DATA SECURITY](https://docs.icme.io/documentation/use-cases/privacy-and-data-security.md) - [Cut your agent observability costs and make every trace auditor-proof](https://docs.icme.io/documentation/use-cases/privacy-and-data-security/cut-your-agent-observability-costs-and-make-every-trace-auditor-proof.md) - [ISO 42001 Enforcement: Evidence Your Auditor Will Actually Accept](https://docs.icme.io/documentation/use-cases/privacy-and-data-security/iso-42001-enforcement-evidence-your-auditor-will-actually-accept.md) - [Protect Your Inbox from Rogue AI Agents](https://docs.icme.io/documentation/use-cases/privacy-and-data-security/protect-your-inbox-from-rogue-ai-agents.md): An AI email assistant that forgets its instructions can delete your entire inbox in seconds. PreFlight makes sure that never happens - even when the agent ignores you. - [HIPAA Patient Data Sharing](https://docs.icme.io/documentation/use-cases/privacy-and-data-security/hipaa-patient-data-sharing.md): This page walks through a complete HIPAA compliance policy for a healthcare AI agent — the policy, the tests, and the key learnings that make it work reliably with automated reasoning. - [Personal Data Access Agent](https://docs.icme.io/documentation/use-cases/privacy-and-data-security/personal-data-access-agent.md) - [Venice AI + Preflight](https://docs.icme.io/documentation/use-cases/privacy-and-data-security/venice-ai-+-preflight.md): Private AI guardrails for private AI inference. Every prompt encrypted, every policy hidden, verifiable decisions that reveal nothing. - [MikeOSS Legal AI](https://docs.icme.io/documentation/use-cases/privacy-and-data-security/mikeoss-legal-ai.md): Open source legal AI document assistant with cryptographic guardrail verification. - [E-COMMERCE](https://docs.icme.io/documentation/use-cases/e-commerce.md) - [Fake Merchant & Phishing Attacks](https://docs.icme.io/documentation/use-cases/e-commerce/fake-merchant-and-phishing-attacks.md): AI shopping agents are completing checkout on fraudulent sites — without the human ever seeing the suspicious domain. - [E-Commerce Cart Hijacking: Prompt Injection Against AI Shopping Agents](https://docs.icme.io/documentation/use-cases/e-commerce/e-commerce-cart-hijacking-prompt-injection-against-ai-shopping-agents.md): Learn how e-commerce cart hijacking attacks manipulate AI shopping agents through hidden product listing instructions, price overrides, quantity changes, and credential exfiltration. - [Cryptographic Guardrails for Your OpenClaw Agent](https://docs.icme.io/documentation/use-cases/cryptographic-guardrails-for-your-openclaw-agent.md): Your OpenClaw agent is sending emails, executing trades, creating accounts, and modifying files — right now, while you're not watching. You have no proof it checked any rule before it acted. - [Guardrails for Self-Evolving OpenClaw Agents](https://docs.icme.io/documentation/use-cases/cryptographic-guardrails-for-your-openclaw-agent/guardrails-for-self-evolving-openclaw-agents.md): How to write and battle test a policy that catches data exfiltration and blocks unauthorized self-modification in OpenClaw agents. - [Blocking ClawHavoc with ICME PreFlight](https://docs.icme.io/documentation/use-cases/cryptographic-guardrails-for-your-openclaw-agent/blocking-clawhavoc-with-icme-preflight.md) ## API Reference - [Developer Platform API](https://docs.icme.io/api-reference/readme.md): Docs for bots.