From a $0.25 verified-financials lookup to a $25 eight-lens institutional report scored across ten classic mental models — the DEALITHIC deal-analysis stack, machine-consumable, priced per call, and every number ships with a receipt.
// your agent, mid-diligence: tool: "sec_extract" args: { ticker: "TSLA" } { "revenue_m": 22387.0, "net_cash_m": 16005.4, "provenance": { "revenue_m": { "tag": "RevenueFromContractWithCustomer…", "form": "10-Q", "filed": "2026-04-23" ← the receipt } } }
Data APIs can tell your agent what the revenue was. So can we — with the filing citation attached. But numbers were never the hard part. The hard part is the judgment that comes after: do these numbers hold up? What breaks under stress? Is this company any good? That's what the flagship returns — and no data feed can follow us there.
Eight analyst lenses on one company, in one call: equity research, investment banking, strategy consulting, turnaround ops, comparables, capital markets, a disclosure audit — and a ten-model judgment framework — synthesized into an executive verdict. Every number cites its SEC filing; anything the filings can't support says "not determinable" instead of guessing.
Authenticate with X-Api-Key. First 25 calls are free on every key.
Verified financial statements and derived metrics for any US-listed filer, straight from EDGAR XBRL — cash, revenue, burn, runway, debt, share count.
The skeptical-senior-analyst pass: tax-rate sanity, EBITDA/EBIT bridge, growth continuity vs history, capex ratio, working-capital stability.
Full underwriting chain: integrity → trend → industry → stress tests → conviction rating, plus a risk register, investment thesis, and rNPV band for biotech.
// an LLM asked for Tesla's debt: "Tesla's total debt is roughly $8–13B" ← a guess // this API: "debt_m": 597.6, "as_of": "2026-03-31", "form": "10-Q", "filed": "2026-04-23" // your compliance team sleeps at night.
The intelligence is our proprietary institutional engine, built on authoritative SEC filings — not a black box you have to send your deals into. It fits your architecture, not the other way around. Your environment. Your compliance posture.
Your UI, your workflow, your storage. You call the engine only for the computation — and most of it runs against public SEC filings, so your confidential deal materials never leave your walls. You ship the product; we're the intelligence underneath.
If nothing can leave your network, run the engine inside your own environment — the analytics engines are self-contained, and the AI reporting runs against your own model credentials. DEALITHIC's capability, entirely on your infrastructure.
FINRA now holds firms accountable for how AI touches their workflows. This is how you adopt it with a posture you can defend — institutional deal intelligence, on your terms.
Hosted MCP means nothing to install — point your client at the URL and your agent has all three tools.
{
"mcpServers": {
"agentic-ai": {
"type": "http",
"url": "https://agentic-ai-production-e711.up.railway.app/mcp",
"headers": { "X-Api-Key": "agai_..." }
}
}
}claude mcp add agentic-ai --transport http \ https://agentic-ai-production-e711.up.railway.app/mcp \ --header "X-Api-Key: agai_..."
pip install dealithic-mcp
{
"mcpServers": {
"agentic-ai": {
"command": "dealithic-mcp",
"env": {
"AGENTIC_AI_API_KEY": "agai_...",
"AGENTIC_AI_BASE_URL": "https://agentic-ai-production-e711.up.railway.app"
}
}
}
}curl -X POST https://agentic-ai-production-e711.up.railway.app/v1/sec/extract \
-H "X-Api-Key: agai_..." -H "Content-Type: application/json" \
-d '{"ticker": "TSLA"}'Tell us what you're building and we'll send your key — same day. After the trial it's pure pay-per-call — $0.25 to $2.00 depending on depth, billed monthly. Stop any time by not calling.
Get my free key