DeSci · Data Infrastructure · Open Standards
Structured mineral
intelligence
AssayChain sources, sanitizes, and delivers mineral data — USGS commodity reports, empirical field studies, and third-party research — as machine-readable API responses with on-chain provenance. Every data point independently verifiable.
Mission
Mineral data should be structured, open — and verifiable
Government mineral data is technically public. But it’s buried in PDFs, scattered across agency websites, and almost impossible to query programmatically. No API. No machine-readable format. No provenance chain linking raw report to delivered data.
Field processing data — recovery rates, feed types, equipment parameters — is almost never shared publicly. Studies happen in isolation. Operators repeat the same experiments because no one structured and published the results from the last time.
AssayChain exists to fix that: source the data, sanitize it against locked schemas, attest it on-chain, and serve it via a pay-per-call API that any AI agent or developer can access without accounts or subscriptions.
DeSci is the movement to rebuild research incentives and data ownership on open infrastructure. In plain terms: instead of locking data in journal paywalls and private databases, you put it on public networks where anyone can verify it. AssayChain applies DeSci principles — on-chain attestations, IPFS storage, permissionless access — to physical-world mineral processing data.
x402 enables machine-to-machine payments without accounts or subscriptions. AI agents can pay for data the same way they pay for any other service call — programmatically, in USDC, with EIP-3009 authorization. No OAuth dance. No API key management. Just pay and get data.
Industry context
What is gravity separation?
The oldest and cleanest form of mineral beneficiation — no chemicals, no heat, just density difference exploited by water flow. Here’s the domain knowledge an AI agent or developer needs to reason about AssayChain data.
Gravity separation
Minerals with different specific gravities separate in a flowing water film on an inclined surface. Dense minerals (gold: SG 19.3, sulfides: SG 4–6) migrate to the concentrate zone. Gangue (quartz: SG 2.65) washes off.
Ultrasound cavitation
A transducer coupled to the tray generates cavitation bubbles at 28–40 kHz. These micro-implosions break surface tension, dislodge fine particles from surfaces, and improve liberation — particularly effective for gold particles below 100 microns.
Placer vs. hardrock
Placer gold is free-milling — already liberated by weathering and erosion, so no crushing required. Hardrock gold is locked inside rock and requires crushing and milling before separation. AssayChain's field research spans multiple mineral processing studies — from placer gravity separation to hardrock feed testing — with each study producing structured, attested run logs.
Recovery rate
The percentage of target mineral actually captured in the concentrate versus what was present in the feed. A recovery rate of 85% means 15% of the gold in the feed was lost to tailings. Every run log records this metric.
Elutriation (example study)
A hydraulic sizing and classification method. In the mini-Duke elutriation setup — one example study currently underway — an upward water column separates fine particles by terminal settling velocity, allowing pre-classification of feed before gravity separation. MineFlowTrack supports any study type, equipment configuration, and processing workflow.
USGS commodity summaries
The U.S. Geological Survey publishes annual Mineral Commodity Summaries covering global mine production, reserve estimates, import/export, price data, and industrial applications. AssayChain serves these as x402-gated, machine-readable API responses with EAS provenance — so AI agents can query USGS data programmatically instead of parsing PDFs.
Ongoing field research
Every study produces structured, attested run logs
MineFlowTrack enables any study type, any equipment, any timeline. The 2026 field season is the current example: runs flow through locked schemas and enter the attestation pipeline regardless of equipment or configuration. Future studies will follow the same pattern.
Current study: Baseline
May 11 – May 19
Grooved stainless tray, no ultrasound, no elutriation. Establishes baseline gravity separation recovery rates for the feed material before any enhancement equipment is introduced.
Schema: tray_geometry: "grooved_stainless_v1"
ultrasound_freq_khz: 0
Current study: mini-Duke
May 20 onward
Mini-Duke elutriation rig added. Feed is hydraulically pre-classified before hitting the tray. Measures whether pre-classification improves recovery rate and reduces processing time.
Schema: tray_geometry: "mini_duke_elutriator_v1"
ultrasound_freq_khz: 0
Current study: Ultrasound
Late June onward
Ultrasound transducer coupled to the grooved stainless tray. Tests whether cavitation at 28 kHz and 40 kHz measurably improves fine gold recovery.
Schema: tray_geometry: "grooved_stainless_v1"
ultrasound_freq_khz: 28 | 40
These are the current 2026 studies. MineFlowTrack supports any future study — different equipment, different feed types, different schemas. The attestation pipeline is study-agnostic.
Who we are
Quetzal Collective LLC
AssayChain is a product of Quetzal Collective LLC — a solo-founded company building structured data infrastructure for the mineral intelligence space. We source data from government agencies (USGS), empirical field studies, and third-party research; sanitize it against locked schemas; and deliver it via a verifiable, pay-per-call API.
Team: a solo operator with a software engineering degree focused on agentic integration (MCP, x402, AI orchestration), working with a PhD geologist consultant for domain validation of extraction outputs, deposit contexts, and compliance-language accuracy. The consultant's domain knowledge is what keeps the geological content defensible.
Data streams are live: USGS Mineral Commodity Summaries for 20 critical minerals, an ongoing empirical field study on gravity separation and ultrasound-assisted recovery, curated US/MX historical mining district reports, daily market context bundles, AI-powered assay extraction from PDFs and images, and company critical-minerals exposure profiles assembled by a multi-agent swarm. Intelligence endpoints (risk, compliance, knowledge-graph, procurement, deposit-model, ore-potential, enrich) synthesize attested data into structured outputs.
Field data is captured using MineFlowTrack — a Quetzal Collective tool built for structured, offline-capable field data entry. MineFlowTrack exports validated run bundles directly into the AssayChain attestation pipeline.
Everything is verifiable: EAS schema UIDs are public, IPFS data is publicly pinned, and the x402 payment gate is mechanical — pay and receive data. No gatekeeping.
Pre-revenue. No analyst desk, no enterprise sales team, no auditor endorsements. Compliance and risk outputs are assistive research context — always require Qualified Person / Competent Person review before any regulatory use. We state this plainly because trust starts with honesty.
Base has the lowest gas costs of any EVM mainnet. EAS (Ethereum Attestation Service) provides a standard schema registry for creating typed, verifiable on-chain records — without a custom smart contract. Both are built for production-scale usage, not experiments.
Storing full run logs directly on the blockchain would cost thousands of dollars in gas fees. Instead, each run is saved on IPFS — a distributed network where a file’s address is derived from its content (change the data, change the address). The EAS attestation holds the permanent on-chain pointer. Anyone can fetch the original data from IPFS and verify it matches what the API returns.
Designed for AI
For AI agents and LLMs
AssayChain ships with Model Context Protocol (MCP) server registration on Anthropic, Glama, and Smithery. Register the MCP endpoint to give any Claude, ChatGPT, or Groq agent access to 19 tools covering the full data ecosystem.
19 MCP Tools +9 new
Free: corpus.search, extract.estimate, sales.ask, benchmark.sample
Preview free / paid via REST: criticality.crosscheck, china_trade_control, supply_share, stockpile_level, supply_resilience, esg_profile, blm_claims_data, geophysics_surveys, active_operations, earth_mri_focus_areas
Paid ($0.10–$0.50): benchmark.commodity, benchmark.runs, district.history, extract.run
No API Keys
Agents pay directly via x402 (USDC on Base). No accounts, no subscriptions, no rate-limit keys to manage. Your agent can call 1 endpoint or 1000; billing is per-call, not per-seat.
On-Chain Receipts
Every paid call returns an x402 receipt — a cryptographically signed proof of payment that can be verified on Basescan. Forward receipts to your auditor. Useful for compliance and audit trails.
Commodity + Compliance Use Cases
Query 20 critical minerals for ESG disclosure, UFLPA sourcing evidence, EU Battery Reg 2023/1542 compliance, CBAM tariff impact, DoD domestic mineral assessments, or M&A due diligence. Agent-friendly schemas with clear examples.
DeepSeek + Groq Powered
Enrichment runs on DeepSeek deepseek-reasoner for deep analysis. Q&A synthesis runs on Groq for sub-2s response. Agents get high-quality answers with inline citations and source attribution.
Persistent Cache — $0.10 Reuse
Extract results and enrichment profiles are now saved to Firestore + IPFS. Subsequent calls for the same source URL or company within 7 days return cached results at $0.10 — 80% cheaper than fresh compute.