Carbon & Cashmere MCP server

Which servers in the official MCP registry are Carbon & Cashmere's own, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.

read 2026-10-26

← Back to Carbon & Cashmere

Whose server it is

Registry namede.carbon-cashmere.api/crypto-intelligenceverified in the MCP registry

The registry name sits under carbon-cashmere.de. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-10-26

Stateanswerschecked 2026-10-26
Tools40
Server calls itselfCarbon & Cashmere
Server version1.27.0
Protocol version2025-06-18

On 2026-10-26 the server answered initialize and tools/list without a sign-in. First answer in 516 ms.

Should an agent connect itNot proven either way yet

Whether its tools do the product's job has not been judged yet.
Checked on 1 of the 7 days it takes to judge reliability.

Once connected

Keep approval on for subtensor_transfer_flow: they delete, send or pay without saying so.
Parameters are untyped or undescribed in 31 of 40 tools: check arguments before calling.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

Measured dailydaily checks: 1, the first on 2026-10-26

INITE MCP scoreafter a weeka score needs 7 daily checks: 1 so far
Answeredmeasuring1 of the 7 daily checks it takes to print a share
Median first answer507 ms
Space in an agent's context≈ 4,151 tokensthe definitions of all its tools (40), which an agent carries on every turn
Tool definitionsB · 87 of 100
Typed results0 of 40 toolsdeclared outputSchema: results come typed, read once

Parameters without a description: in 31 of 40

What an agent can do with it

not sure · check daily market regime and on-chain flow summaries
not sure · track large Bitcoin on-chain movements through live alerts
can do · pull crypto market data into bots via API

Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?

Worth a look

Tools that delete or pay, marked read-only: subtensor_transfer_flow

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (30%), doing the product's job (25%), definitions (20%), context (10%) and use (15%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

Tools it offers40 tools · read 2026-10-26

kaspa_toccata_status
Kaspa Toccata hardfork tracker: live activation distance, ETA computed from public Kaspa node data (api.kaspa.org), post-fork parameter reference. Informational research data - not investment advice. $0.05
kaspa_fee_regime
Kaspa fee-market regime: fee-bucket distribution incl. the sub-floor segment the Toccata minimum displaces. Informational research data - not investment advice. $0.05
kaspa_throughput_health
Kaspa throughput & DAG-width: chain-block rate, estimated total DAG rate, TPS vs multi-month pre-fork baseline - hardfork impact tracking. Informational research data - not investment advice. $0.10
kaspa_timeseries
Historical Kaspa network metric as aggregated timeseries (max 1000 points). Metrics: hashrate, difficulty, block_rate, mempool_tx_count, peer_count (only until 2026-08-10), daa_score, tip_count, node_hashrate, circulating_supply…
bittensor_subnet_stake_flows
dTAO stake-flow microstructure per subnet: per-neuron accumulation/distribution, net flow, concentration trend from months of continuous high-cadence snapshots. Informational research data - not investment advice. $0.10
bittensor_subnet_coordination
Neutral Bittensor subnet concentration & synchrony metrics: stake/emission concentration, validator dominance, update-synchrony with 7d comparison. Informational research data - not investment advice. $0.10
btc_miner_attribution
Bitcoin mining-pool hashrate share derived from coinbase scriptSig parsing. Returns block-count + share % per pool for window (1h/24h/7d/30d), with explicit Unknown bucket for long-tail small pools. $0.03
btc_ordinals_flow
Hourly Bitcoin Ordinals inscriptions + OP_RETURN byte volume. Inscriptions detected via witness-envelope magic bytes (cord). Window up to 7d. $0.02
btc_coin_days_destroyed
Hourly Bitcoin Coin Days Destroyed proxy (consolidation_ratio × fee_weight × volume) + whale-tx contribution. ⚠️ PROXY methodology - suitable for trend signals, not absolute reserve-coin-age accounting. $0.03
btc_diversity
Bitcoin own-node peer-network diversity (IPv4/IPv6/Tor/I2P/CJDNS distribution) + privacy share. AI agent API for privacy researchers, node-operator tooling, network-health analytics. Informational research data - not investment advice…
btc_mempool_pressure
Bitcoin mempool pressure classification (low/medium/high/extreme) + fee buckets (0-1, 1-2, 2-5, 5-10, 10-20, 20-50, 50-100, 100-200, 200+ sat/vB) + RBF count + CPFP candidate count. Updated every 5 min. $0.02
btc_fee_curve
Bitcoin smart-fee curve for 1/3/6/12/24/144-block confirmation targets + trend classification (rising/stable/falling/surging) from 2h moving average. $0.02
btc_blocks
Deep per-block Bitcoin extract. Returns miner-pool attribution, witness/taproot adoption %, Ordinals inscription count, OP_RETURN byte volume, fee percentiles (10/25/50/75/90), UTXO delta, segwit %, total output BTC, block weight. $0.02
btc_blocks_range
Bulk Bitcoin block extract (max 144 blocks per call, flat price per call). Returns array of per-block summaries (height, time, miner pool, tx count, witness/taproot/ordinals/op_return stats, total output, coinbase reward). $0.02
btc_coinbase_messages
Curated Bitcoin coinbase scriptSig messages (printable ASCII only). Pool tags, BIP-comment campaigns, political statements. Window up to 7d. $0.02
kaspa_dag_state
Kaspa BlockDAG topology snapshot: tips, sink, virtual chain, pruning point, blue scores. For DAG-state monitoring, side-branch detection, sync verification. $0.02
kaspa_chain_density
Kaspa chain-blocks (selected-parent chain) vs total DAG blocks per hour - measures BlockDAG branching activity. Chain ratio drift signals DAG-mode shifts. $0.05
kaspa_supply_trend
Kaspa circulating supply over time + emission rate + percent-mined. Captures accelerating geometric halving schedule. AI agent API for tokenomics tracking and supply-emission models. Informational research data - not investment advice…
kaspa_mempool_pulse
Kaspa mempool transaction count over time. AI agent API for mempool-state regime detection. Informational research data - not investment advice. $0.02
kaspa_blocks
Single Kaspa block extract with full DAG topology (parents, selected-parent, blue/daa score, chain-block flag). AI agent API for DAG-aware indexers. Informational research data - not investment advice. $0.02
kaspa_recent_blocks
Recent Kaspa blocks (max 144 blocks per call, flat price per call). Optional chain_only flag to filter selected-parent chain blocks. AI agent API for backtesters and indexers. Informational research data - not investment advice. $0.02
subtensor_registration_pulse
Bittensor NeuronRegistered count per hour. Tracks new neurons joining subnets - useful for detecting subnet growth/decay and registration-burn pressure. AI agent API for subnet-health analysts. Informational research data - not investment…
subtensor_transfer_flow
Bittensor BalancesTransfer count + total TAO volume per hour. AI agent API for TAO whale-watch and treasury-movement monitoring. Informational research data - not investment advice. $0.05
subtensor_weights_activity
Bittensor WeightsSet + TimelockedWeightsCommitted/Revealed activity per hour. Tracks validator weight-setting cadence and the post-2025 anti-frontrun mechanism. AI agent API for validator-behavior analysts. Informational research data…
subtensor_block_metrics
Recent per-block aggregation of Bittensor events: neuron_registrations, weight_sets, bond_sets, child_key_sets, stake_added/removed, root_claimed, balance_transfers + total TAO. $0.02
subtensor_evm_activity
EVM-on-Bittensor Ethereum.Executed call count per hour. Tracks TaoEVM dApp adoption. AI agent API for cross-VM adoption analysts. Informational research data - not investment advice. $0.05
subtensor_events
All Bittensor substrate events for a single finalized block (event_idx, phase, module_id, event_id, attributes). AI agent API for detailed forensic analysis of specific blocks. Informational research data - not investment advice. $0.02
bittensor_subnet
Per-subnet Bittensor snapshot: static metadata (modality/owner/burn/kappa/rho) + latest Tier-A metrics + current alpha-token price. Covers all 129 mainnet subnets. $0.02
bittensor_subnet_metagraph
Full metagraph snapshot for 13 tracked Bittensor subnets [1,3,4,13,15,18,26,56,62,64,91,120,123]: summary stats + all neurons with stake/incentive/dividends/emission/trust/validator_permit. $0.05
bittensor_subnet_validators
Top-N validators ranked by incentive for tracked Bittensor subnets. Filter only validator_permit=true neurons. AI agent API for validator-selection and stake-delegation decisions. Informational research data - not investment advice. $0.02
bittensor_subnet_emission_trend
Hourly emission and stake trend for a Bittensor subnet (window: 1h/24h/7d/30d). Derived from bittensor_metagraph snapshots. AI agent API for subnet-momentum tracking. Informational research data - not investment advice. $0.02
bittensor_subnets_all
Snapshot of all 129 Bittensor mainnet subnets - burn, neurons, owner, modality, alpha-price. For cross-subnet research and market-overview dashboards. $0.05
bittensor_alpha_prices
Latest alpha-token price per Bittensor subnet (TAO + USD). All 129 subnets in one call. AI agent API for cross-subnet price arbitrage, dTAO portfolio rebalancing. Informational research data - not investment advice. $0.02
bittensor_alpha_price_history
Time-series of alpha-token price for a single Bittensor subnet (window: 1h/24h/7d/30d). Sourced from bittensor_alpha_prices captured every 10 minutes. AI agent API for alpha-token trading signals. Informational research data - not…
bittensor_network_fundamentals
Latest TAO network-wide fundamentals: total issuance, total stake, free float, supply percentage mined, stake ratio. Snapshot every 10 min. $0.02
bittensor_registration_cost_history
Hourly Bittensor subnet registration activity (NeuronRegistered event count per subnet) + current burn-cost per subnet. Useful for predicting cheap-registration windows. $0.02
bittensor_top_validators_cross_subnet
Top-N validators per tracked Bittensor subnet, aggregated across all 13 focus subnets. One call returns the leaderboard across subnets. For cross-subnet validator-research and stake-allocation decisions. $0.05
bittensor_yield_history
Realized Bittensor validator yield per subnet measured from continuous snapshots (30d series + per-validator APR) - not theoretical APY. $0.25
forecast_archive
Matured AI forecast archive with realized outcomes - verifiable benchmark dataset of directional and cross-sectional crypto forecasts across major assets. Strictly retrospective. Call the scorecard for the list of covered forecast tracks…
forecast_scorecard
Rolling verifiable AI-model quality scorecard on matured forecasts, with disclosed random baselines and standard classification metrics. $0.25

How to connect

Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:

{
  "mcpServers": {
    "crypto-intelligence": {
      "type": "http",
      "url": "https://api.carbon-cashmere.de/mcp/"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http carbon-cashmere https://api.carbon-cashmere.de/mcp/

Add to Cursor Add to VS Code

claude.ai, ChatGPT and other clients: add https://api.carbon-cashmere.de/mcp/ as a custom connector in the settings.

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.