Evidenced brief · 4 findings · credit-first
The world's first enhanced-weathering offtake, a genuinely farm-distributed operating model, zero agent-discovery surface, and the field-labor angle on partner farms.
No jargon. Lithos's world-first offtake and farm-distributed model are real and credited; the fixes below are about the one gap this review found: nothing an agent can read. Here is what each one is worth.
Nothing on lithoscarbon.com is machine-readable to an agent, so an AI sourcing credible removal has nothing to read; a simple agent surface puts your real, first-of-its-kind track record in front of the tools now doing that sourcing.
Your $57M world-first offtake and farm-distributed model live only as prose; expressed as machine-readable fact, an AI describes what makes Lithos structurally different instead of lumping you in with industrial-site suppliers.
Credit-first: real capability and honesty are named before any gap. Every claim below carries a plain reverify command: run it yourself.
$57.1M for 154,240 tonnes over 2024–2028, an implied $370/tonne, announced Dec 2023: Frontier's own writeup names this the world's first enhanced-weathering offtake purchase.
curl -s https://frontierclimate.com/writing/lithos | grep -oE '\$[0-9.]+M|[0-9,]+ tonnes|first enhanced weathering'Neither /llms.txt (the file AI assistants read first) (4,326B, matches a root nonsense-path control exactly) nor /.well-known/agent-card.json (the file that lets an AI assistant use a site correctly) (88B, matches a /.well-known/-prefixed nonsense control exactly) exists: both are clean true 404s, no soft-200 deception, just nothing published.
curl -s -o /dev/null -w "%{http_code}:%{size_download}\n" https://www.lithoscarbon.com/llms.txt https://www.lithoscarbon.com/xyzzy-control-2026.txt https://www.lithoscarbon.com/.well-known/agent-card.json https://www.lithoscarbon.com/.well-known/xyzzy-control-2026.jsonUnlike most carbon-removal suppliers we've reviewed, Lithos's removal happens directly on active, working agricultural land (an ag-co-benefit model) rather than at a dedicated industrial site: a structurally different field-operations footprint.
curl -s https://www.lithoscarbon.com/ | grep -io 'farm\|basalt\|weathering' | sort | uniq -cBasalt spreading and verification sampling across many partner farms is inherently scattered, single-visit field work: the exact shape EWP packet-prices (fund→claim→prove→settle) rather than a fixed on-site crew.
EWP's own fund → claim → prove → settle lifecycle (internal working notes)Lithos Carbon does not need EcoWealth: nothing here claims a relationship, a gap in their operations, or a pitch. What follows is a concept: the bounded, photo+GPS+signature-proven field tasks EWP already prices per-packet, mapped onto the kind of physical plant/field work an enhanced weathering operation runs day to day.
Concept only: not a proposal to Lithos Carbon, not a claim they lack this capability internally, and not evidence any EWP packet has ever been posted for this pathway. EWP's own proof (photo+GPS+one-time signature) is a task-completion layer, not a carbon-quantification or registry-issuance layer: durable-CDR crediting requires accredited MRV far heavier than a single field packet, as EcoWealth's own Carbonfuture recon (2026-07-16) already found in detail.
