Skip to main content
Glama
134,951 tools. Last updated 2026-05-14 13:33

"Hyper" matching MCP tools:

  • Build a complete Tableau dashboard from a CSV file (end-to-end). Pipeline: CSV → schema inference → chart suggestion → Hyper extract → workbook creation → chart configuration → dashboard layout → .twbx output. IMPORTANT FOR AI AGENTS — know before you call: * Auto-selected charts are picked by RULES (data shape + heuristics), NOT by natural-language chart requests. To guarantee a specific chart (e.g. "top 10 customers by profit") you MUST pass it via ``required_charts``; otherwise it may or may not be generated and may be trimmed out by ``max_charts``. * The user's reference image is NOT automatically read from the chat. If the user asks for styling "like this image", you must save the image to disk and pass its path via ``reference_image``. * This tool returns a STRUCTURED MANIFEST dict describing what was actually built. Always cite fields from the returned manifest (worksheets / theme / dashboards) when describing the result to the user — do NOT invent charts that are not listed there. Args: csv_path: Path to the source CSV file. output_path: Output .twbx path (defaults to <csv_stem>_dashboard.twbx). dashboard_title: Dashboard title (derived from filename if empty). max_charts: Maximum number of charts (0 = use dashboard_rules.yaml default). template_path: TWB template path (empty for default template). theme: Theme preset name (empty = use dashboard_rules.yaml default). Options: modern-light, modern-dark, classic, minimal, vibrant. rules_yaml: Optional YAML string with dashboard rules overrides. Example: "kpi:\n font_size: 32\n max_kpis: 3" required_charts: Optional list of chart specs that MUST be built. Each entry is a dict, for example:: {"title": "Top 10 Customers by Profit", "kind": "bar", "rows": "Customer Name", "columns": "SUM(Profit)", "top_n": 10, "top_by": "SUM(Profit)", "sort_descending": "SUM(Profit)"} Supported kinds: bar, line, scatter, pie, map, heatmap, tree_map, text. Required charts bypass dedup + trim. reference_image: Optional path to a PNG/JPG whose palette should drive the dashboard styling. Applied AFTER the theme so image colours win. The path must exist on disk; chat attachments need to be saved first. Returns: Structured manifest dict with keys: ``status``, ``output_path``, ``source``, ``dashboards``, ``worksheets``, ``charts_built``, ``required_charts_fulfilled``, ``dropped_suggestions``, ``theme``, ``warnings``, ``summary``. Cite THIS dict when describing results.
    Connector
  • Get Hyperliquid perpetual futures trader leaderboard with performance metrics. Returns: Trader performance rankings as markdown. Columns returned: - **Address**: Trader's wallet address - **Label**: Nansen label of the trader (if available) - **Total PnL**: Total profit/loss in USD (currency formatted, can be negative) - **ROI**: Return on investment as percentage (percentage formatted) - **Account Value**: Total account value in USD (currency formatted) **Sorting and Filtering Options**: You can sort and filter (from/to amounts) on these fields: totalPnl, accountValue, roi Example: ``` { "date": {"from": "7D_AGO", "to": "NOW"}, "accountValue": {"from": 100000, "to": 1000000}, "totalPnl": {"from": 10000}, "order_by": "totalPnl", "orderByDirection": "DESC" } ``` Notes: - Hyperliquid-specific endpoint (perpetual futures only)
    Connector
  • Build a complete Tableau dashboard from a Hyper extract file (end-to-end). Pipeline: Hyper → schema inference → chart suggestion → workbook creation → chart configuration → dashboard layout → .twbx output. IMPORTANT FOR AI AGENTS: auto-charts come from rules, not from natural-language requests. Use ``required_charts`` to guarantee a specific chart and ``reference_image`` for image-based styling. The returned manifest dict is the SOURCE OF TRUTH — don't invent charts that aren't in it. Args: hyper_path: Path to the .hyper file. output_path: Output .twbx path (defaults to <hyper_stem>_dashboard.twbx). dashboard_title: Dashboard title (derived from filename if empty). max_charts: Maximum number of charts (0 = use rules default). template_path: TWB template path (empty for default template). table_name: Table name inside the Hyper file (empty = first table). theme: Theme preset name (empty = use rules default). rules_yaml: Optional YAML string with dashboard rules overrides. required_charts: See ``csv_to_dashboard.required_charts``. reference_image: See ``csv_to_dashboard.reference_image``. Returns: Structured manifest dict describing what was actually built.
    Connector
  • Profile the currently connected data source. Works for ANY connection type: CSV extract, Hyper, MySQL, Tableau Server, Excel — anything that has fields in the workbook. The profile includes dimension/measure classification, semantic types, domain hints, and boolean signals that guide chart and template selection. Args: source_type: Override source detection. Usually "auto" which inspects the workbook fields. Other options: "csv", "hyper" (requires separate file path tools). Returns: Human-readable DataProfile with signals for template/chart decisions.
    Connector
  • Get Hyperliquid perpetual futures trader leaderboard with performance metrics. Returns: Trader performance rankings as markdown. Columns returned: - **Address**: Trader's wallet address - **Label**: Nansen label of the trader (if available) - **Total PnL**: Total profit/loss in USD (currency formatted, can be negative) - **ROI**: Return on investment as percentage (percentage formatted) - **Account Value**: Total account value in USD (currency formatted) **Sorting and Filtering Options**: You can sort and filter (from/to amounts) on these fields: totalPnl, accountValue, roi Example: ``` { "date": {"from": "7D_AGO", "to": "NOW"}, "accountValue": {"from": 100000, "to": 1000000}, "totalPnl": {"from": 10000}, "order_by": "totalPnl", "orderByDirection": "DESC" } ``` Notes: - Hyperliquid-specific endpoint (perpetual futures only)
    Connector
  • Inspect a Hyper extract file and return its schema with column classification. Reads the Hyper file, maps column types, classifies columns as dimensions or measures, and returns a summary. Requires tableauhyperapi (pip install tableauhyperapi). Args: hyper_path: Path to the .hyper file. table_name: Specific table to inspect (empty = first table). Returns: Human-readable schema summary with dimensions, measures, and types.
    Connector

Matching MCP Servers

Matching MCP Connectors

  • Generate hyper-personalized cold email sequences via MachFive API.

  • Publish complete HTML artifacts and return a public share URL.

  • Convert a CSV file to a Tableau Hyper extract. Infers column types and creates a .hyper file that can be used as a data source in Tableau workbooks. Requires tableauhyperapi (pip install tableauhyperapi). Args: csv_path: Path to the source CSV file. hyper_path: Output path for the .hyper file. table_name: Table name inside the Hyper file. sample_rows: Rows to sample for type inference. Returns: Confirmation with row and column counts.
    Connector