# Pronto Intelligence Fabric — Commercial Data Dictionary & Specification > **Version**: 2.5.0 (Enterprise Production Specification) > **Published by**: Ovasabi Studios / Pronto > **Contact**: > **Companion Dataset**: `pronto_signals_sample_1000.csv` --- ## 1. Executive Summary The **Pronto Signal Fabric** is an enterprise-grade, high-throughput intelligence platform that ingests, cleans, harmonizes, and cross-correlates multi-domain real-time data from over 200 authoritative global sources. In ~3.5 months of continuous production operations (launched mid-July 2026), Pronto has: * **Processed over 541,219,000 signal events** (370.1M+ multilingual RSS & breaking geopolitical feeds, 171.0M+ structured market, crypto, and macroeconomic signals). * **Handled over 667,000 edge visitor and API requests** across 180+ sovereign countries and 75 localized languages. * Operated a continuous **Enterprise Fusion Layer**: an auditable suite of 35 peer-reviewed quantitative models that compute real-time composite indices (triangular cross-rate arbitrage, macro stress, recession probability, grid carbon dislocation, and multi-hazard convergence). Pronto serves as the foundational multi-modal bridge between live financial, geopolitical, and physical market real-world observations and autonomous AI decision agents. --- ## 2. Verified Institutional & Machine Demand Profile Pronto’s high-density data surfaces are continuously indexed, scraped, and dialed by frontier artificial intelligence labs, search infrastructure, and autonomous agent networks. Our durable edge counters confirm massive, verified consumption: * **Anthropic (`ClaudeBot`)**: **269,615 requests** systematically harvesting localized hubs, macro stress indices, and multi-lingual country data. * **Meta (`Meta-ExternalAgent`)**: **56,958 requests** tracking real-time topics and breaking event feeds. * **Amazon (`Amazonbot`)**: **48,545 requests** indexing deep regional indicators, macro resilience, and fire/climate hazard surfaces across 75 languages. * **Huawei (`Petalbot`)**: **25,756 requests** indexing supply chain and cyber exploitation risk models. * **Autonomous Agent Networks & MCP Registries**: * **83,260 MCP discovery activations** (`initialize`, `server/discover`, and `tools/list`). * **10,506 direct automated agent calls** via Model Context Protocol from runtimes like AgentRank, mcpbeat, and TaifoonHarvester. * **Other Frontier AI Indexers**: Recurring verified sessions from OpenAI (`GPTBot`, `OAI-SearchBot`), Perplexity (`PerplexityBot`), DeepSeek (`DeepSeekBot`), Alibaba (`QwenBot`), Mistral AI, and ByteDance (`ByteSpider`). --- ## 3. Dataset Architecture & Schema The companion evaluation file `pronto_signals_sample_1000.csv` is a stratified sample of **exactly 1,000 production records** structured under a unified 14-field tabular specification. ### Field Definitions | Field Name | Type | Nullable | Description | Sample Value | | :--- | :--- | :--- | :--- | :--- | | `signal_id` | String | No | Unique deterministic identifier for the signal or fusion instance. | `pronto_fuse_crypto_cross_rate_dislocation_001` | | `layer` | Enum | No | Architecture tier: `ground_truth_signal` (direct observational stream) or `enterprise_fusion` (derived quantitative model). | `enterprise_fusion` | | `category` | String | No | High-level asset class or thematic domain. | `Crypto & Digital Assets`, `Macroeconomics`, `Prediction Markets` | | `product_or_collection` | String | No | Underlying database collection or fusion model machine identifier. | `crypto_cross_rate_dislocation`, `prediction`, `macro`, `crypto` | | `provider_or_model` | String | No | Primary authoritative upstream source or quantitative calculation engine. | `Coinbase`, `Manifold`, `World Bank`, `Pronto Triangular Arbitrage Engine` | | `target_entity_or_asset` | String | No | Ticker, currency pair, contract title, sovereign country, or CVE incident ID. | `BTC/USD`, `US Treasury Term Structure`, `CVE-2026-0024` | | `published_at_utc` | Timestamp | No | Canonical ISO 8601 UTC timestamp of observation or calculation. | `2026-10-24T15:18:17Z` | | `headline` | String | No | Human-readable title, market contract outcome, or concise indicator reading. | `Triangular cross deviation for BTC/EUR vs BTC/USD: 14.8 bps` | | `observation_details` | String | Yes | Contextual payload: order book depth, price velocity, moving averages, or formula citations. | `Bid: 68420.50 \| Ask: 68422.00 \| 24h Vol: $1.2B` | | `primary_value` | Float | No | Headline numeric value normalized for quantitative modelling. | `0.00148`, `0.3140`, `68421.25`, `7.25` | | `value_unit` | String | No | Unit of measurement for `primary_value`. | `USD`, `Probability (0-1)`, `Basis Points`, `Percent` | | `severity_or_tier` | Enum | No | Urgency ranking (`INFO`, `LOW`, `MEDIUM`, `HIGH`, `CRITICAL`) or licensing tier (`COMMERCIAL`, `ENTERPRISE`). | `COMMERCIAL`, `HIGH`, `CRITICAL` | | `confidence_score` | Float | No | Input completeness, freshness, and sensor consensus metric (0.00 to 1.00). | `0.98`, `1.00` | | `citation_or_grounding` | String | No | Traceable grounding: peer-reviewed literature citation, upstream venue handle, or standard ID. | `Cassel (1918); Levich Ch. 3`, `worldbank:AG.LND.FRST.ZS` | --- ## 4. The Enterprise Fusion Layer (Catalogue of 35 Models) Unlike black-box algorithmic scores, every product in Pronto's Enterprise Fusion Layer is a **faithful implementation of a published, citable scientific or economic formula**, making all outputs fully defensible, auditable, and reproducible: ### Macroeconomics & Monetary Policy * **`recession_risk_12m` (US Recession Risk 12-Month)**: Implements the Estrella & Mishkin (1998) probit equation over the 10-year minus 3-month Treasury yield spread: $P(R) = \Phi(-0.5333 - 0.6379 \times \text{spread})$. * **`macro_stress_index` (Macro Stress Index)**: Integrates Arthur Okun's Misery Index (headline CPI + unemployment), the Fisher (1930) exact real rate, and the 2s10s yield curve slope. * **`sahm_recession_indicator` (Sahm Rule Recession Onset)**: Implements Claudia Sahm’s (2019) rule tracking real-time 3-month moving average unemployment divergence from the 12-month low. * **`taylor_rule_policy_gap` (Taylor Rule Policy Gap)**: Computes the policy spread between actual central bank target rates and John Taylor's (1993) benchmark rule: $R = r^* + \pi + 0.5(\pi - \pi^*) + 0.5(y - y^*)$. * **`sovereign_debt_dynamics` (Sovereign Debt Sustainability)**: Evaluates sovereign $(r - g)$ growth-interest differentials following the Olivier Blanchard (2019) framework. * **`fx_reference_rate_consistency` & `fx_fixing_drift`**: Evaluates continuous interbank liquidity against official daily fixings (ECB, BoE, WMR 4pm). ### Digital Assets & Quantitative Arbitrage * **`crypto_cross_rate_dislocation` (Triangular Currency Dislocation)**: Evaluates the law of one price (Cassel 1918; Levich Ch. 3) by comparing continuous cross-exchange crypto-fiat pairs against central bank reference fixings to measure liquidity fragmentation and basis opportunities. * **`insider_filing_cluster` (Abnormal Insider Accumulation)**: Conviction-weighted scoring of open-market corporate Form 4 insider transactions (Lakonishok & Lee 2001), filtering routine 10b5-1 plans and option grants. ### Energy, Climate & Physical Hazards * **`grid_carbon_arbitrage` (Cross-Border Interconnection Carbon Spread)**: Real-time marginal emissions factor differentials across synchronous interconnectors (ENTSO-E). * **`compound_hazard_convergence` (Multi-Hazard Copula Convergence)**: Joint probabilistic hazard estimation (Zscheischler et al., Nature Reviews 2020) across seismic, meteorological, and grid reserves. * **`fire_danger_index` (Canadian Forest Fire Weather Index - FWI)**: Computes fuel moisture codes (FFMC, DMC, DC) and rate of fire spread (Van Wagner 1987). * **`seismic_population_exposure` (USGS PAGER Framework)**: Real-time convolution of ShakeMap peak ground acceleration (PGA) against LandScan global population grids (Wald et al. 2008). * **`cyber_exploitation_risk` (Weaponized Vulnerability Cluster)**: Harmonizes CVSS v3.1, FIRST.org Exploit Prediction Scoring System (EPSS), and CISA Known Exploited Vulnerabilities (KEV). --- ## 5. Delivery Methods & Technical Specifications | Parameter | Ground-Truth Signals | Enterprise Fusion Layer | | :--- | :--- | :--- | | **Total Processed Volume** | **541,219,000+** events processed | Continuous real-time recalculation | | **Total Retained State** | **28,269,000+** indexed historical records | 35 published quantitative models | | **Edge Ingress History** | **667,000+** requests served across ~100 days | Zero latency on hotplane memory | | **Ingestion Velocity** | ~15,000 to 45,000 structured events/day | Real-time on every signal window tick | | **Supported Protocols** | Model Context Protocol (MCP / Streamable HTTP), REST JSON, WebSocket Hotplane, Kafka stream | Model Context Protocol (MCP tools), CWF v3 compact wire, REST JSON | | **Geographic Coverage** | 180+ sovereign nations, 75 localized languages | Global, regional, and event-level scopes | | **Entity Grounding** | Wikidata QIDs, ISO-3166, ISO-4217, SEC CIK, LEI | Grounded with source correlation handles | | **Privacy & Compliance** | **100% de-identified**: zero personal data, zero user credentials, zero internal execution tables | Strictly non-PII, mathematically auditable | --- ## 6. Commercial Licensing & Access Tiers 1. **Evaluation Tier (Sample Pack)**: * 1,000 representative records (`pronto_signals_sample_1000.csv`) with full schema rights for internal testing and feasibility analysis. 2. **Standard Commercial Feed**: * Continuous live streaming access via Streamable HTTP (MCP) or WebSocket hotplane across selected collections (Crypto, Prediction, Macro, or Equities). 3. **Enterprise Full-Spectrum Feed**: * Complete historical archive (multi-hundred-million lifecycle records) delivered via compressed Parquet objects in S3/MinIO, paired with dedicated real-time Kafka event streams and unrestricted access to all 35 Enterprise Fusion products.