---
name: api.carbon-cashmere.de
description: api.carbon-cashmere.de provides ML-generated crypto trading signals, market anomaly detection, sentiment-tagged news aggregation, and live network health data for Bitcoin, Kaspa, and Bittensor (TAO). It combines proprietary conviction scoring and directional signal feeds with infrastructure monitoring endpoints, all served as single-shot REST queries. No order execution, streaming, or price feed primitives are offered.
host: api.carbon-cashmere.de
---

# api.carbon-cashmere.de

Carbon Cashmere is a signal and analytics host targeting crypto trading agents that need pre-processed, model-derived intelligence rather than raw market data. It serves agents making allocation or trade-evaluation decisions by supplying directional signals, anomaly flags, conviction scores, and news sentiment in a unified API surface. Its distinguishing characteristic is the combination of on-chain network health checks (Kaspa BlockDAG, Bittensor TAO) alongside ML signal layers, making it useful for agents that monitor both market conditions and underlying network fundamentals.

## When to use this host

Use api.carbon-cashmere.de when an agent needs pre-processed, model-derived trading signals, anomaly flags, or conviction scores for crypto assets — particularly BTC — without needing to run its own ML pipeline. It is also the right host when an agent must monitor Kaspa or Bittensor TAO network health alongside market signals in a single API surface. Do not use this host for real-time price quotes, OHLCV candles, order book data, or trade execution — route those needs to a dedicated market data provider (e.g. CoinGecko, Binance, or a CEX API). Do not use it for historical network data, wallet balances, subnet-level Bittensor metrics, or signals on assets outside the supported coin list. The news endpoint covers crypto only; for general financial news use a broader news API.

## Capabilities

### Directional Trading Signals

Provides ML-generated long/short/neutral signals and market alerts across crypto assets, giving agents actionable directional bias with probability and confidence metadata before trade evaluation.

- **`fetch-carbon-signal-feed`** — Returns ML-generated long/short/neutral trading signals with probability, confidence, and optional track record for up to 20 cryptocurrencies.
- **`fetch-carbon-market-signals`** — Returns recent crypto market signals from Carbon Cashmere, filterable by signal type, severity, symbol, lookback window, and result count.

### Bitcoin Conviction Analytics

Delivers a composite BTC-specific conviction score (0–100) with directional bias and a breakdown of contributing signal components, enabling agents to make or justify Bitcoin allocation decisions.

- **`fetch-btc-conviction-score`** — Returns a composite conviction score (0–100) for Bitcoin combining consensus, market extremity, confluence, threshold breaches, and ML anomaly signals.

### Market Anomaly Detection

Flags unusual market patterns across crypto assets using ML-scored anomaly events, including RSI deviations, volume spikes, and funding rate anomalies, with severity and feature-level breakdowns.

- **`fetch-market-anomalies`** — Returns ML-detected market anomaly signals across crypto assets, including anomaly score, severity, and feature breakdown per detected event.

### Crypto News and Sentiment

Aggregates recent crypto news articles from multiple sources with per-article sentiment labels and coin tags, allowing agents to incorporate market narrative context into decisions.

- **`fetch-crypto-news`** — Returns aggregated crypto news articles from CryptoCompare and RSS feeds, with sentiment scoring, coin tags, and category labels, for a configurable lookback window.

### Network Health Monitoring

Returns live infrastructure snapshots for the Kaspa BlockDAG and Bittensor TAO networks, covering block height, hashrate, peer count, sync status, and stake metrics.

- **`fetch-kaspa-network-stats`** — Returns live Kaspa BlockDAG network stats including block count, hashrate, mempool size, peer count, and sync status from a self-hosted Rusty Kaspa node.
- **`fetch-tao-network-data`** — Returns current Bittensor TAO network metrics including block number, peer count, sync status, token metadata, and total stake in TAO.

## Workflows

### BTC Trade Conviction Pipeline

*Use when an agent needs to build a multi-layered case for or against a Bitcoin position by combining directional signals, anomaly context, and news sentiment before acting.*

1. **`fetch-btc-conviction-score`** — Retrieve the composite BTC conviction score and directional bias as the primary decision anchor.
2. **`fetch-carbon-signal-feed`** — Pull the ML directional signal for BTC to confirm or challenge the conviction score direction and probability.
3. **`fetch-market-anomalies`** — Check whether any active anomaly events on BTC (volume spikes, RSI extremes, funding deviations) explain or undermine the signal.
4. **`fetch-crypto-news`** — Fetch recent BTC-tagged news with sentiment to surface any narrative catalyst that aligns with or contradicts the signal direction.

### Altcoin Signal with Anomaly Validation

*Use when an agent needs to evaluate a directional signal for a non-BTC crypto asset and wants to verify it is not driven by an anomalous market event before acting.*

1. **`fetch-carbon-signal-feed`** — Retrieve the ML long/short/neutral signal and probability for the target altcoin.
2. **`fetch-market-anomalies`** — Check whether the asset appears in current anomaly detections that could distort the signal's reliability.
3. **`fetch-carbon-market-signals`** — Query recent market alerts (liquidation cascades, confluence events, correlation breaks) for the same symbol to add event-level context.

### Kaspa Infrastructure and Signal Check

*Use when an agent needs to assess whether Kaspa network health supports or undermines a KAS trading signal before making an infrastructure or trading decision.*

1. **`fetch-kaspa-network-stats`** — Retrieve live Kaspa BlockDAG metrics including hashrate, mempool size, peer count, and sync status.
2. **`fetch-carbon-signal-feed`** — Pull the ML directional signal for KAS to see whether network conditions align with the market signal direction.

## Skill reference

### `fetch-carbon-signal-feed`

**Carbon Signal Feed** — Returns ML-generated long/short/neutral trading signals with probability, confidence, and optional track record for up to 20 cryptocurrencies.

*Use when:* Use when an agent needs actionable directional trading signals (long/short/neutral) with conviction scores and market regime context for one or more crypto assets before placing or evaluating a trade.

*Not for:* Do not use for real-time price feeds or order book data; use a market data API instead. Not suitable for fetching signals on assets outside the supported coin list.

**Inputs:**

- `coins` (string) — Comma-separated list of coin tickers to filter results (e.g. BTC,ETH,SOL). Defaults to all coins with available signals.
- `actionable_only` (boolean) — When true, returns only coins with actionable signals (probability >= 0.6 and confidence >= 0.15). Defaults to false.

**Returns:** Returns count, actionable_count, market_regime, and a signals map keyed by coin ticker, each with direction, probability, confidence, actionable flag, source, regime, and optional 30-day backtest track record.

**Example:** `GET https://api.carbon-cashmere.de/v1/signal-feed?coins=BTC,ETH,SOL&actionable_only=false`

---

### `fetch-kaspa-network-stats`

**Kaspa Network Check** — Returns live Kaspa BlockDAG network stats including block count, hashrate, mempool size, peer count, and sync status from a self-hosted Rusty Kaspa node.

*Use when:* Use when an agent needs current Kaspa network health data such as hashrate, blocks per second, mempool size, peer count, or sync status to monitor the Kaspa BlockDAG or inform trading/infrastructure decisions.

*Not for:* Do not use for Kaspa price or market data; use a market data endpoint instead. Not suitable for historical network data — this returns a single live snapshot only.

**Returns:** Returns a JSON object with block_count, header_count, difficulty, hashrate_thash, mempool_size, peer_count, is_synced, timestamp, source, and blocks_per_second from a live Rusty Kaspa node.

**Example:** `GET https://api.carbon-cashmere.de/v1/kaspa/network`

---

### `fetch-crypto-news`

**Carbon Cashmere Crypto News** — Returns aggregated crypto news articles from CryptoCompare and RSS feeds, with sentiment scoring, coin tags, and category labels, for a configurable lookback window.

*Use when:* Use when an agent needs recent crypto news articles optionally filtered by coin symbol (e.g. BTC, ETH, SOL), with sentiment labels and source URLs, to inform trading decisions or summarize market narrative.

*Not for:* Do not use for real-time price feeds or order book data; use a market data endpoint instead. Not suitable for non-crypto general news.

**Inputs:**

- `coin` (string) — Filter results by coin symbol (e.g. BTC, ETH, SOL). If omitted, returns news across all coins.
- `hours` (integer) — Lookback window in hours. Minimum 1, maximum 168. Defaults to 24.
- `limit` (integer) — Maximum number of articles to return. Minimum 1, maximum 100. Defaults to 30.

**Returns:** Returns a count integer and an articles array where each item has a title, source, sentiment label (neutral/positive/negative), coin symbol tags, category tags, article URL, and ISO 8601 timestamp.

**Example:** `GET https://api.carbon-cashmere.de/v1/news?coin=BTC&hours=24&limit=10`

---

### `fetch-market-anomalies`

**Anomaly Pulse** — Returns ML-detected market anomaly signals across crypto assets, including anomaly score, severity, and feature breakdown per detected event.

*Use when:* Use when an agent needs to identify unusual market patterns across crypto assets, such as abnormal RSI, volume spikes, price changes, or funding rate deviations flagged by an ML model.

*Not for:* Do not use for real-time streaming price feeds or standard OHLCV data; use a market data or price feed API instead. Not suitable for retrieving normal price quotes without anomaly context.

**Returns:** Returns an anomalies array where each entry contains an integer id, ISO timestamp, asset symbol, ML anomaly score, severity label, a features object with RSI/volume/price/funding metrics, and a text description.

**Example:** `GET https://api.carbon-cashmere.de/v1/anomalies`

---

### `fetch-btc-conviction-score`

**BTC Conviction Score** — Returns a composite conviction score (0–100) for Bitcoin combining consensus, market extremity, confluence, threshold breaches, and ML anomaly signals.

*Use when:* Use when an agent needs a single actionable Bitcoin market conviction score with directional bias (bullish/neutral/bearish) and breakdown of contributing signal components before making a trading or allocation decision.

*Not for:* Do not use for real-time price feeds or order execution; this is a single-shot analytical score, not a live ticker or trade API. Not suitable for coins other than BTC on this endpoint.

**Inputs:**

- `coin` (string, required) — Ticker symbol of the coin to score; this endpoint is fixed to BTC as a path parameter.

**Returns:** Returns coin, integer score (0–100), tier, direction, a components object with five weighted sub-scores (consensus, market_extremity, confluence, threshold_breaches, ml_anomaly), and an updated_at timestamp.

**Example:** `GET https://api.carbon-cashmere.de/v1/conviction/BTC`

---

### `fetch-tao-network-data`

**TAO Network Data** — Returns current Bittensor TAO network metrics including block number, peer count, sync status, token metadata, and total stake in TAO.

*Use when:* Use when an agent needs live Bittensor network state such as current block height, number of peers, whether the node is syncing, or total staked TAO amount.

*Not for:* Do not use for historical TAO price data, wallet balances, or subnet-level metrics; use a dedicated Bittensor indexer or exchange API instead.

**Returns:** Returns a JSON object with coin symbol, chain name, peer count, sync status, block number, token metadata, total staked TAO, data source, and last-updated timestamp.

**Example:** `GET https://api.carbon-cashmere.de/v1/network/TAO`

---

### `fetch-carbon-market-signals`

**Carbon Signals** — Returns recent crypto market signals from Carbon Cashmere, filterable by signal type, severity, symbol, lookback window, and result count.

*Use when:* Use when an agent needs recent market alerts or trading signals for specific crypto assets or across all assets, including liquidation cascades, yield shifts, confluence events, correlation breaks, and BTC dominance changes.

*Not for:* Do not use for real-time streaming price feeds or order execution; this is a single-shot paginated signal query, not a live data stream or trade router.

**Inputs:**

- `type` (string) — Filter by signal type (e.g. liquidation_cascade, yield_shift, confluence, correlation_break, btc_dominance).
- `severity` (string) — Filter by severity level: low, medium, high, or critical.
- `symbol` (string) — Filter by asset symbol (e.g. BTC, ETH, TON) or pool identifier.
- `hours` (integer) — Lookback window in hours; minimum 1, maximum 168, default 24.
- `limit` (integer) — Maximum number of signals to return; minimum 1, maximum 200, default 50.

**Returns:** Returns a count integer and a signals array where each entry includes id, ts, signal_type, symbol, severity, and a data object with type-specific fields such as price, change_24h, liquidation amounts, APY shifts, or confluence details.

**Example:** `GET https://api.carbon-cashmere.de/v1/signals?symbol=BTC&severity=critical&hours=48&limit=10`

---
