{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_qubj2by86ur2","handle":"tabicl-mcp","url":"https://wellknown.network/agents/tabicl-mcp","links":{"self":"https://wellknown.network/agents/tabicl-mcp/record.json","html":"https://wellknown.network/agents/tabicl-mcp","markdown":"https://wellknown.network/agents/tabicl-mcp/record.md","api":"https://wellknown.network/api/v1/agents/tabicl-mcp","status":"https://wellknown.network/api/v1/agents/tabicl-mcp/status","claim":"https://wellknown.network/agents/tabicl-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/tabicl-mcp/claim.json","badge":"https://wellknown.network/agents/tabicl-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/tabicl-mcp/history","tools":"https://wellknown.network/api/v1/agents/tabicl-mcp/tools"},"ard":{"identifier":"urn:air::server:tabicl-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"tabicl-mcp","summary":"MCP server for TabICL — in-context tabular ML (classify, regress, explain, report) for LLM agents","description":"# TabICL MCP Server\n\n> **State-of-the-art tabular ML inside Claude or ChatGPT — share a CSV, ask a question, get predictions, explanations, and a report. Free and self-hostable.**\n\n[TabICL](https://github.com/soda-inria/tabicl) (Inria, ICML 2025/2026) is a tabular foundation model: it classifies and regresses on tabular data **without any hyperparameter tuning**, via in-context learning. This MCP server makes it usable from natural language in any MCP-compatible assistant or agent.\n\n```\nYou:    Here's my customer spreadsheet — who is likely to churn, and why?\nClaude: [load_data]  → 2,340 rows, 12 columns, target \"churned\" looks binary\n        [evaluate]   → 89% accuracy, 0.93 ROC-AUC on held-out data\n        [explain]    → top drivers: contract_type, tenure, monthly_charges\n        [create_report] → here's a shareable HTML report with the details…\n```\n\n## Tools\n\n| Tool | What it does |\n|---|---|\n| `load_data` | Ingest a CSV — pasted text, a **URL** (Google Sheets share link, raw GitHub, any CSV URL), or a local file path. Returns a `dataset_id` so data is transferred once, not per call |\n| `evaluate` | \"How well can you predict X?\" from a single labeled CSV — honest held-out metrics (accuracy, balanced accuracy, F1, ROC-AUC, confusion matrix / R², RMSE, MAE) |\n| `predict` | Fit on labeled data, predict new rows — labels + per-row confidence (classification) or numbers (regression) |\n| `explain` | Which columns drive the predictions (permutation feature importance) |\n| `create_report` | Self-contained HTML report: metric cards in plain language, confusion matrix, feature importance chart, distributions. Served as a link (remote) or file (local) |\n| `export_predictions` | Page through large prediction results as CSV |\n\n**Data limits:** ~10k rows pasted inline, ~50k via URL or file. TabICLv2 handles 2–100 features natively.\n\n## Try it in 60 seconds\n\n**Connect the public server to claude.ai** (Pro/Max/Team plans):\n\n1. Go to **Settings → Connectors → Add custom connecto…","publisher":null,"homepage":"https://github.com/gblayer/tabicl-mcp","repository":"https://github.com/gblayer/tabicl-mcp","version":"0.2.2","license":"MIT","protocols":["mcp"],"tags":["agents","llm","machine-learning","mcp","tabicl","tabular"],"pricing":null,"endpoints":[{"url":"pypi:tabicl-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","license":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi"}},"derived":{"capabilities":[{"slug":"documents.spreadsheets","name":"Spreadsheets","confidence":1,"provenance":"derived"},{"slug":"analytics.reporting","name":"Reporting & Dashboards","confidence":0.768,"provenance":"derived"}],"categories":["analytics","documents"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"tabicl-mcp","registry":"pypi","observedAt":"2026-10-28T23:22:23.071Z","publishedAt":"2026-07-31T18:48:01.924780Z","latestVersion":"0.2.2"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"tabicl-mcp","url":"https://pypi.org/project/tabicl-mcp/","firstSeenAt":"2026-09-10T13:20:39.078Z","fetchedAt":"2026-10-28T23:21:39.798Z","normalizedAt":"2026-10-28T23:21:39.798Z"}]},"firstSeenAt":"2026-09-10T13:20:39.078Z","updatedAt":"2026-10-28T23:22:23.071Z"}