IRL puts a cryptographic seal on every autonomous trading decision before it reaches the exchange. Pre-execution policy enforcement, bitemporal audit ledger, post-trade verification. Bring your own model and signal, or use MacroPulse as a turnkey Market Truth Anchor.
Current observability infrastructure captures what autonomous agents did and when. It cannot prove why — or what the agent knew at the moment of decision. This is the gap that regulators are asking about, and that no existing tool closes.
| Capability | Current state | With IRL |
|---|---|---|
| What action was taken | Logged | Logged + sealed |
| When the action occurred | Single timestamp | Bitemporal (valid_time + txn_time) |
| Why the action was taken | Approximate / inferred | Deterministic CognitiveSnapshot |
| What the agent knew at decision time | Not provable | Cryptographically proven via L_t fingerprint |
| Whether constraints were enforced | Not guaranteed | Pre-execution hard halt on violation |
| Which model produced the decision | Not verifiable | Model hash in Multi-Agent Registry |
| Whether execution matched reasoning | Not tracked | Post-trade verifier with divergence detection |
Every trade produces a cryptographic chain from regime attestation to exchange confirmation. No party can alter any link without breaking the chain.
reasoning_hash + trace_id — embedded in exchange order metadata before submission.exchange_tx_id.final_proof = SHA-256(reasoning_hash ‖ exchange_tx_id) — chain closed.Three demo agents are pre-seeded. Use Swagger UI to run a full authorize → bind-execution cycle in-browser. No API key, no registration.
Each layer is independently deployable. Start with L1 in under a day. Upgrade when your regulatory requirements escalate.
IRL depends on an abstract MTA interface, not on any specific operator. Any cryptographically signed regime source can serve as the MTA for your deployment.
Zero additional infrastructure. MacroPulse's HMM/PCA pipeline produces Ed25519-signed regime broadcasts with version hashes. Plug in your URL and public key — IRL verifies signatures and caches state automatically. Recommended for firms without a proprietary regime model.
Quant firms running their own regime pipelines implement a single Rust trait (MtaClient) and pass it at startup. IRL seals against your signal with the same cryptographic guarantees. Your model stays private. Your compliance chain is identical.
IRL produces evidence that maps directly to specific regulatory requirements — not general "auditability," but the specific proof each framework asks for.
| Requirement | IRL mechanism |
|---|---|
| Decision auditability — EU AI Act Art. 12 | reasoning_hash = SHA-256(canonical snapshot) |
| Transparency to regulators — Art. 13 | ZK proof: compliance verified without alpha disclosure |
| Pre-trade risk controls — SEC Rule 15c3-5 | Policy Engine blocks non-compliant intents before execution |
| Contemporaneous recordkeeping — MiFID II | Bitemporal timestamps: valid_time + txn_time, immutable |
| Fleet governance — DORA, AI Act | Multi-Agent Registry: per-agent identity and risk controls |
Compliance value is in the seal existing — not in counting seals. L1 is operational in under a day. Upgrade when your regulatory requirements escalate.
Volume discounts — L1 and L2
| Fleet size | Discount |
|---|---|
| 1–5 agents | List price |
| 6–15 agents | −15% |
| 16–30 agents | −25% |
| 31+ agents | −35% (negotiated) |
Custom MTA deployments (proprietary signal) pay standard L1/L2/L3 pricing — no additional fee for signal-agnostic configurations. MacroPulse MTA as an L1 add-on: $300/month.
The API is live. Three demo agents are pre-seeded. Try authorize + bind-execution in Swagger UI before committing to a deployment conversation.
gabriel.veron134@gmail.com