Services

Project mindshare time series.

Get OpenAI-compatible inference from many models at market prices via Surplus Intelligence.

Send a question about longevity, aging, or lifespan science. Returns a request ID to poll for the AI-generated answer from Aubrai, VitaDAO's decentralized co-scientist.

LLM-generated narrative recap of recent prediction-market activity.

Filter wallets by market trades with AND/OR logic

List Polymarket events with filtering and sorting

Get the global leaderboard of smart wallets

OT-contextualised CVE triage for ICS/SCADA environments. Pass ?id=CVE-XXXX-XXXX. Returns OT-adjusted severity (recalculated from NVD CVSS for the affected ICS layer), cyber-physical impact category, patch feasibility without downtime, CISA KEV status, and a prioritised recommended action. Enriched by DeepSeek LLM for ICS-specific context. Every call returns fresh data from NVD, CISA-KEV, and DeepSeek.

SCADA/ICS internet-exposed device lookup for industrial control systems. Pass ?vendor=siemens&model=s7-1200. Returns default credential risk, at-risk OT protocols (Modbus/502, DNP3/20000, S7comm/102), exploitation notes, and hardening action. Covers Siemens, Schneider, Rockwell, Honeywell, GE, Unitronics, Beckhoff.

IOC enrichment with ICS/OT campaign context. Pass ?value=1.2.3.4&type=ip (or type=domain). Queries AlienVault OTX (reputation, pulse count, OT-tagged feeds) + AbuseIPDB (abuse confidence score, ISP, Tor node detection) + DeepSeek CTI analysis for known ICS threat actor association and campaign context. Returns structured verdict on whether the IOC is linked to OT-targeting campaigns. Cheapest endpoint at $0.01.

OT/ICS patch feasibility assessment for a CVE ID. Pass ?id=CVE-XXXX-XXXX. Returns patch availability (from NVD references), vendor advisory URLs, OT-safe workarounds, patch complexity for the affected ICS layer, estimated downtime in minutes, whether safe to patch live, recommended maintenance window, deployment strategy, CISA KEV status, and a risk-vs-disruption score (1–10 with rationale). DeepSeek-enriched.

ICS threat actor profile lookup. Pass ?name=SANDWORM. Fetches live data from MITRE ATT&CK for ICS (STIX bundle) + CISA ICS advisories, enriched by DeepSeek. Returns: MITRE technique mappings (ICS-specific T-codes), known malware/tools, related groups, recommended OT detections, attribution, physical impact assessment, and last known activity. Works for any known ICS threat actor — not limited to a fixed list. Alias lookup supported (e.g. Volt Typhoon → VOLTZITE, APT44 → SANDWORM).

Query Polymarket markets with filtering, sorting, and pagination

Classify BIM model level of detail against LOD standards

Fetch trading activity (merges, splits, redeems) for a user

Calculate compliance training completion rates by role

Calculate optimal model retraining trigger thresholds

Detect jailbreak and prompt injection attempt patterns

Reference for LLM benchmarks: MMLU, HumanEval, GSM8K, and LMSYS Arena.

Calculate maximum sustainable yield from population model data

Raw SQL query against 80+ ClickHouse tables (sub-second).

Latest Stock Trends Inference Model (ST-IM) outputs for a symbol: forward return expectations and statistical distributions across 4-week, 13-week, and 40-week horizons. Use /v1/meta/stim for ST-IM provenance, base-period context, and interpretation limits.

Token price history.

List sports markets grouped by game

Send an HPKE-encrypted prompt to the Aubrai chat agent running inside a Tinfoil TEE. Only the enclave can decrypt your message.

Discover markets where high-performing wallets are active

Aggregated wallet profile across chains.

Fear & Greed index history.

Token rankings (market cap, volume, change).

Name in, official IDs out. Give it a company or crypto name or ticker ("Apple", "ETH", "Stripe") and get the verified IDs that identify it everywhere: SEC CIK + ticker, GLEIF LEI, Wikidata QID, CoinGecko slug. Every ID is pulled live from the entity's own official registry and tagged with the sources that agreed — so you can check it yourself, not trust an LLM's guess. Ranked by confidence.

Bundled market state inference API: combines BTC/ETH regime direction, liquidation cascade risk + open interest velocity, and composite crypto sentiment in a single call. Discounted vs $0.07 a-la-carte. Designed for trading agents that need a complete market snapshot per decision cycle.

Fetch + clean a web page into markdown.

Fetch any URL and return clean Markdown with metadata and LLM-ready chunks — flat $0.10 per call, no credit multiplier

Free LLM inference — Llama 3.3, Gemma 3, Mistral, Qwen via a single endpoint

Technical indicators (RSI, MACD, Bollinger, EMA).

Query historical Polymarket trade data with flexible filtering

Prioritize radiology images by clinical urgency using AI detection of critical findings in X-rays, CT, and MRI scans. Re

Yeetful — Anthropic Claude Haiku 4.5 inference over MCP Streamable HTTP, hosted at anthropic.yeetful.com. Exposes ask_claude (single-prompt completion) and claude_chat (multi-turn) tools, capped at 256 output tokens per call. Pay-per-call in USDC on Base. Operated by yeetful.com. Keywords: yeetful, anthropic, claude, haiku, mcp, x402, inference, llm.

Live web data enricher data for agents and LLMs.

Get OHLCV candlestick data for a Binance trading pair

Count tokens and estimate LLM costs for 15+ models before every API call — built for AI agents

Live web data quality scorer data for agents and LLMs.

AI-curated crypto news feed.