ICME Preflight
Jailbreak-proof guardrails for AI agents.
What it can do
- Check Logic: FREE — no account needed. Check if agent reasoning is logically consistent. Extracts quantitative and logical claims, then uses a Z3 SAT solver to mathematically prove whether they contra
- Create Account: Create an ICME Preflight account via x402 USDC payment ($5.00 on Base). Returns an API key and 325 starting credits. Save the api_key — it is shown only once.
- Create Account Card: Create an ICME Preflight account via credit card ($5.00). Returns a checkout_url to open in a browser. After payment, call poll_session with the session_id to get your API key.
What data it sees
Do you need an account
No: the server works without sign-in
Jailbreak-proof guardrails for AI agents. Policy enforcement powered by Automated Reasoning and formal verification — an SMT solver, not an LLM, decides whether an action passes or fails. Cannot be prompt-injected. Every decision produces a cryptographic ZK proof.
Includes a FREE check_logic tool that catches contradictions in agent reasoning (budget overflows, impossible timelines, conflicting constraints) using a Z3 SAT solver. No account needed.
13 tools covering the full workflow:
check_logic — FREE. Mathematically prove reasoning is consistent before acting on it. make_rules — write guardrails in plain English, ICME compiles them to formal logic via Automated Reasoning. check_action / quick_check — verify any agent action against your policy. SAT = allowed, UNSAT = blocked. verify_proof — independently verify the ZK receipt from any prior check. get_scenarios / run_tests — test your policy with AWS Automated Reasoning scenarios before deploying. Account & billing — create account ($5 USDC or card), top up credits, pay per call via x402.
Docs: https://docs.icme.io
Server tool list (13)
Raw names from tools/list. Only developers need these.
| check_logic | FREE — no account needed. Check if agent reasoning is logically consistent. Extracts quantitative and logical claims, then uses a Z3 SAT solver to mathematically prove whether they contradict. Returns CONSISTENT, CONTRADICTION, or UNKNOWN with the extracted claims. Use this before acting on any plan, calculation, or chain of thought. |
| create_account | Create an ICME Preflight account via x402 USDC payment ($5.00 on Base). Returns an API key and 325 starting credits. Save the api_key — it is shown only once. |
| create_account_card | Create an ICME Preflight account via credit card ($5.00). Returns a checkout_url to open in a browser. After payment, call poll_session with the session_id to get your API key. |
| make_rules | Compile a natural language policy into formal logic (SMT-LIB). Costs 300 credits. Write your guardrail in plain English (up to 50 rules). Returns a policy_id and scenarios for review. Requires X-API-Key. |
| check_action_paid | Verify an agent action against a guardrail policy via x402 payment ($0.10 USDC on Base). No API key needed. Returns SAT (allowed) or UNSAT (blocked) with extracted values and optional ZK proof. Tip: end action with an explicit claim. |
| check_action | Verify an agent action against a guardrail policy (1 credit). Cheapest path if you have credits from signup or top_up. Requires X-API-Key. Returns SAT/UNSAT with extracted values and ZK proof. |
| verify_action | Minimal verification: returns ALLOWED or BLOCKED (1 credit). Requires X-API-Key. Use when you only need the verdict. |
| verify_proof | Verify a ZK proof from a prior check. No additional cost. Wait a few minutes after the check for the proof to be ready. Single-use per proof. |
| top_up | Add 500 credits via x402 USDC payment ($5.00 on Base). Requires X-API-Key to identify the account. |
| top_up_card | Add credits via credit card. Returns a checkout_url. Tiers: $5=500, $10=1050, $25=2750, $50=5750, $100=12000 credits. Requires X-API-Key. |
| poll_session | Poll a Stripe Checkout session after card payment. Returns pending/complete status. For signup, returns the api_key. |
| get_scenarios | Get generated test scenarios for a policy. Requires X-API-Key. |
| run_tests | Run saved test cases against a policy. Requires X-API-Key. All tests should pass before production use. |