Crypto sentiment API

The core endpoint is sentiment-scored article search. Every piece of content in the index carries a sentiment label (positive, neutral, or negative) assigned by an AI model at ingestion, plus detected entities with their own per-entity sentiment. Query by keyword, company, outlet, language, region, or date range and get back structured JSON with sentiment breakdowns, top outlets, and per-outlet average tone.

On top of raw search, the API serves aggregated signals: historical sentiment time series by outlet, the Fear & Greed Index, AI-extracted narrative trends with signal strength, narrative momentum scores that tell you whether a theme is accelerating or fading, and side-by-side entity comparisons across volume, sentiment, and sources.

The distinction that matters: this is media sentiment. The index covers news outlets, financial press, crypto publications, podcasts, conference talks, regulatory filings, and GitHub activity. It is the layer of conversation institutions actually read.

Generate a key from Settings, then authenticate with a Bearer token:

curl "https://api.perception.to/feed?keyword=bitcoin&sentiment=negative" \
  -H "Authorization: Bearer pcp_your_key_here"

Or skip the code entirely. The same key connects Perception to Claude, ChatGPT, and Gemini over MCP, where the sentiment tools become native functions your AI assistant calls mid-conversation. One-click installs exist for Claude Desktop, VS Code, and Cursor. See the MCP integration page for setup.

Rate limit headers ship on every response, keys are revocable instantly, and the free key tier needs no credit card. Full endpoint reference in the documentation.

Trading dashboards. Pull daily sentiment and Fear & Greed into your existing panel next to price and funding. Divergence between media tone and price is the signal most builders end up alerting on.

Research notebooks. Query historical sentiment by outlet and date range into pandas, and backtest how coverage tone led or lagged price at previous cycle turns.

Comms monitoring. Track sentiment on your own company and competitors, broken down by outlet, so you know which stories moved tone and which outlets to brief next.

AI agents. Connect over MCP and let your assistant run the whole loop: search coverage, compare entities, check momentum, and draft the brief. No glue code.