MCP servercom.crowdcent/challenge
AI agents build, backtest and run ML prediction models: market data, cloud GPUs, live trading.
Overview
Score?
UNRATED 0.143
of what a free look can see, on 1 looks
Looks
2
last 5 hr ago
Tools
15
More info
URL
mcp.crowdcent.com/mcp
streamable-http
Says it is
crowdcent 3.4.11
protocol 2025-06-18
In the record since
4 days ago
Among servers32,723 with a card
0median 0.547 · this server 0.143 · highest on record 0.8351
Toolsfrom sha256:13a402e4c7…68e950
| Tool | Schema |
|---|---|
| blend_simulations Blend up to 5-50 weighted sleeves (tier-capped: 5 Challenger, 10
Contender, 25 Centurion, 50 Sovereign) into one ensemble
book; returns blend stats plus the sleeve correlation matr |
input · output |
| get_challenge_info Get detailed information about a challenge. |
input · output |
| get_inference_data_info Get details for one inference period ('current' works). |
input · output |
| get_inference_data_url Get a signed, time-limited download URL for inference features. |
input · output |
| get_meta_model_url Get a signed, time-limited download URL for the consolidated
meta-model file. |
input · output |
| get_performance Get your historical submission performance (scores flattened per
row, newest first). |
input · output |
| get_submission Get details for one of your submissions by id. |
input · output |
| get_training_dataset_info Get details for one training dataset version. |
input · output |
| get_training_dataset_url Get a signed, time-limited download URL for a training dataset.
Works everywhere (no local filesystem needed): fetch the URL with
any HTTP client. |
input · output |
| list_all_challenges List all active CrowdCent challenges.
Use this to discover challenge slugs before calling
challenge-scoped tools. |
input · output |
| list_submissions List your prediction submissions for a challenge. |
input · output |
| list_training_datasets List all training dataset versions for a challenge. |
input · output |
| run_simulation Backtest one portfolio config against the live meta-model.
The simulator trades the meta-model's published rankings as a
long/short portfolio with your chosen construction (cohort |
input · output |
| submit_predictions_from_dataframe Submit predictions from a JSON representation of a dataframe.
The dataframe must contain the challenge's required columns (for
hyperliquid-ranking: id, pred_10d, pred_30d). |
input · output |
| sweep_simulations Grid-search up to your tier's budget (96 configs at Contender,
24 below) of portfolio constructions in one call.
IMPORTANT: choose candidates on in-sample numbers (`is_stats`),
pr |
input · output |
Verify it yourself
npx teppi-check https://mcp.crowdcent.com/mcpcurl -s https://api.teppi.xyz/v1/trust/mcp/mcs_01M7C9KW21JYGKG00SQZPZHWHA