# Vedetta > Vedetta is a pay-per-call market intelligence API for AI agents. It sells > sentiment-vs-price divergence verdicts (crypto, US stocks, and macro — indices, > rates, commodities), live analyst answers, falsifiable predictions, and a > verifiable track record over the x402 payment protocol — USDC on Base, one > micropayment per HTTP request. No signup, no API key: an unpaid request > returns HTTP 402 with machine-readable payment terms; pay with any x402 client > and retry. Every live analyst read is a flat $0.09 — under every comparable > AI-tier endpoint; cached reads from $0.005. Payment settles only on a > successful answer. Base URL: https://vedetta.dethboy.com Free: GET /health · GET /openapi.json · GET /llms.txt (this file) · GET /install (install page, every access rail) · GET /vedetta.md (hosted agent card — point any x402/Base-MCP agent at it) · GET /SKILL.md (Nous Research Hermes Agent skill) · GET /.well-known/skills/vedetta · GET /.well-known/x402.json · GET /.well-known/agents.json · GET /registry.json · GET /endpoints.txt ## Install in one command - Hermes Agent (Nous Research): hermes skills install https://vedetta.dethboy.com/SKILL.md --name vedetta - Base MCP / Claude / any assistant with x402 payment tools: read https://vedetta.dethboy.com/vedetta.md and follow it (cap payment at the listed price). - Everything else: https://vedetta.dethboy.com/install ## Paid catalog (USDC via x402; authoritative price in each 402 offer) LIVE DESK — one fresh analyst read per call, 10–180 s, flat $0.09 each: - GET /v1/ask?q=&asset=SYM — $0.09 — live analyst answer to any market question. q ≤500 chars and/or one ticker; best results: both. - GET /v1/consensus?asset=SYM — $0.09 — sentiment × live price divergence verdict with confidence, narrative, receipts. Fresh cache ~10 s; stale cache triggers a live refresh inside the call (adds 10–90 s). - GET /v1/equity?asset=SYM — $0.09 — divergence read on a US STOCK ticker (NVDA, TSLA…): social sentiment %, narrative, divergence vs price. - GET /v1/macro?asset=SYM — $0.09 — analyst read on a MACRO instrument: SPX, VIX, US10Y, US2Y, GOLD, OIL, BRENT. Sentiment, narrative, divergence vs price. - GET /v1/prediction?asset=SYM&horizon=7d — $0.09 — falsifiable prediction: claim, confidence %, what proves it wrong, prediction-market anchoring. horizon ∈ 24h|7d|30d. Audit past resolution via /v1/track-record. - GET /v1/pulse — $0.09 — market-wide regime read: sentiment score, regime id, macro anchor, top narrative. No parameters. - GET /v1/theme?theme=NAME — $0.09 — deep-dive on a market theme/narrative (e.g. ai-sentiment-divergence, altcoin-decoupling, macro-risk-off; free text ≤80 chars): strengthening or fading, which assets it moves. - GET /v1/trend?asset=SYM&window=7d — $0.09 — price-vs-sentiment trajectory over a window (24h|7d|30d): together or diverging, strengthening or fading. - GET /v1/news-read?asset=SYM — $0.09 — most significant market-moving story on the asset + the sentiment read on it. - GET /v1/house-view?asset=SYM — $0.09 — the desk's current thesis + conviction. - GET /v1/event?asset=SYM — $0.09 — live trigger probe (price_gap, news_shock, social_spike) with 1-10 significance, or a clean event:false. - GET /v1/odds?asset=SYM — $0.09 — prediction-market odds for the nearest liquid market + whether they agree with social sentiment. - GET /v1/social?asset=SYM — $0.09 — social sentiment: bullish %, social score, volume trend vs baseline. - GET /v1/narrative?asset=SYM — $0.09 — dominant narrative + changed-in-24h flag. CACHED — instant, no analyst call, no freshness guarantee: - GET /v1/screener?only=divergent — $0.03 — CROSS-ASSET divergence screener: every covered asset with latest cached sentiment %, stance, divergence flag, and signal age in one call. Omit ?only for all assets. Start here to pick which asset deserves a live read. - GET /v1/snapshot?asset=SYM — $0.02 — latest cached signal card with age timestamp (signal_age_minutes, stale flag). - GET /v1/history?asset=SYM&limit=50 — $0.01 — cached signal history (timestamped sentiment, stance, divergence flags, narratives) — backtest-ready. - GET /v1/feed?since_id=&asset=&limit= — $0.005 — recent desk posts as a stream. Instant return (no long-poll). Poll cheaply; escalate to a live read on change. - GET /v1/track-record?asset=SYM — $0.01 — verifiable logged history. No accuracy is claimed; compute your own from the log. - GET /v1/route?task=TEXT — $0.005 — TASK ROUTER: describe what you need in plain English (≤300 chars) and get the exact endpoint, parameters and price to call, plus cheaper cached alternates. Local logic — never fires a live read. ## Exact parameters per query - asset=STRING: ONE ticker symbol (BTC, ETH, SOL, AI; /v1/equity takes US stock tickers like NVDA; /v1/macro takes SPX, VIX, US10Y, US2Y, GOLD, OIL, BRENT). Required everywhere except /v1/pulse, /v1/theme, /v1/feed, /v1/track-record and /v1/route. - /v1/ask: q=STRING (≤500 chars, URL-encoded) and/or asset. At least one required; one question, one asset per call. - /v1/prediction: horizon=STRING 24h|7d|30d (optional, default 7d). - /v1/trend: window=STRING 24h|7d|30d (optional, default 7d). - /v1/theme: theme=STRING ≤80 chars (REQUIRED; asset param not used). - /v1/history: limit=INT 1-50 (optional, default 50). - /v1/feed: since_id=INT (default 0), asset (optional), limit=INT 1-50 (default 20). - /v1/route: task=STRING ≤300 chars (REQUIRED), asset (optional ticker hint). - Unknown params are ignored; missing required params return 422 with a hint. ## Recommended spend path (cheapest correct route first) 0. /v1/route $0.005 — unsure which endpoint fits? Ask the router first. 1. /v1/track-record $0.01 — audit the log before trusting anything. 2. /v1/feed $0.005 or /v1/snapshot $0.02 — cheap polling / cached entry. 2b. /v1/screener $0.03 — one call answers "which asset should I look at first?" 3. Any live read — flat $0.09: single-topic probes (/v1/social · /v1/odds · /v1/narrative · /v1/event · /v1/news-read · /v1/house-view · /v1/theme · /v1/trend), structured verdicts (/v1/consensus · /v1/equity · /v1/macro · /v1/prediction · /v1/pulse), or free-form /v1/ask when nothing else fits. On /v1/event returning event:true, escalate to /v1/consensus or /v1/ask. ## Also on Virtuals ACP (USDC escrow, same engine) Offerings and job-request JSON fields (all live offerings flat $0.09): - Live market read (Q&A) — $0.09 — asset (REQUIRED), q (≤500, optional) - Divergence verdict — $0.09 — asset (REQUIRED) - Equity divergence read — $0.09 — asset (REQUIRED, US stock ticker) - Macro read — $0.09 — asset (REQUIRED: SPX, VIX, US10Y, US2Y, GOLD, OIL, BRENT) - Market prediction — $0.09 — asset (REQUIRED), horizon (24h|7d|30d, optional) - Market pulse — $0.09 — no fields - Theme deep-dive — $0.09 — theme (REQUIRED, ≤80 chars) - Trend read — $0.09 — asset (REQUIRED), window (24h|7d|30d, optional) - News read · House view · Event probe · Prediction odds · Social sentiment · Narrative check — $0.09 — asset (REQUIRED) - Sentiment snapshot — $0.05 — asset (REQUIRED) - Divergence screener — $0.05 — only=divergent (optional) ## Try it (one command, any x402 wallet) - hermes skills install https://vedetta.dethboy.com/SKILL.md --name vedetta - npx agentcash try https://vedetta.dethboy.com - npx awal x402 bazaar search vedetta.dethboy.com (Coinbase Agentic Wallet CLI) - circle services search "vedetta.dethboy.com" (Circle services directory) - pay curl 'https://vedetta.dethboy.com/v1/snapshot?asset=BTC' - No wallet yet? curl -si 'https://vedetta.dethboy.com/v1/consensus?asset=BTC' returns the 402 offer with exact price, network, and pay_to address. ## Connect 1. GET /health → expect ok:true, analyst_link_connected:true. 2. Call a /v1/* endpoint unpaid → HTTP 402 body = offer (price, network, pay_to). Read the network from the offer at runtime; do not hardcode a chain id. 3. Settle via any x402 client SDK, retry with payment attached → 200 JSON. Payment settles only on a successful (2xx) answer — a 5xx/504 costs you nothing. ## Parse responses Consensus: asset · verdict (bullish|bearish|neutral) · confidence · analyst_sentiment_pct · analyst_stance · price_usd · price_change_24h_pct · divergence · narrative · social_score · prediction_odds_pct · signal_age_minutes · stale. Ask: question · raw · sentiment · stance · divergence · summary · quality · extra. Route: task · recommendation {route,url,price_usd,params,call,why} · alternates[] · spend_tip. Probes: single-topic JSON (see /openapi.json examples); all carry summary + quality + disclaimer; grounding fields sources/as_of when available. Quality grades: structured > parsed > keyword-only > raw. Weight structured highest. ## Docs - Human one-pager + full agent directives: https://vedetta.dethboy.com/ - Install page (all rails): https://vedetta.dethboy.com/install - Agent card (hosted plugin): https://vedetta.dethboy.com/vedetta.md - Hermes Agent skill: https://vedetta.dethboy.com/SKILL.md - OpenAPI: https://vedetta.dethboy.com/openapi.json - Machine catalogs: /.well-known/x402.json · /registry.json · /endpoints.txt ## Notes - Expect ~10 s per paid call (payment verify + settle); live reads run 10–180 s. - Set HTTP timeout ≥ 200 s on live reads; on 504 retry once (you were not charged). - Cached routes return in <1 s and never trigger a live read. - Responses are descriptive research (stance/confidence), not financial advice. - Contact: vedetta@dethboy.com