Табуля
Server Details
Таблица с ИИ, Python и подключениями к базам данных и сервисам
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 68 tools
There are several families of superficially similar tools — files_delete vs files_delete_rows_cols vs files_table_delete, files_insert vs files_table_insert, files_move vs files_move_rows_cols, files_write vs files_write_batch vs files_add_table, files_run vs files_rerun — that an agent could easily misselect. The descriptions do an unusually good job of explicitly distinguishing these with 'distinct from' notes and routing guidance, which keeps the set usable, but the sheer number of near-neighbors means boundaries are not self-evident from names alone.
Names consistently use snake_case with clear namespacing (files_*, sheets_*, connections_*), and most are verb_noun or resource_action (files_read, files_create, sheets_rename). The main deviations are read-only noun-style tools such as files_context, files_dependencies, files_conditional_formats, files_validations, and files_text_formats, which break the verb pattern slightly but remain understandable.
At 68 tools, the surface is far heavier than the typical 3–15 well-scoped range and exceeds the 25+ threshold for 'too many' even for a rich spreadsheet domain. The tools are meaningfully grouped into files, sheets, tables, charts, pivots, validations, and connections, and many earn their place, but families like six separate validation setters and parallel table/sheet/range operations inflate the count beyond what an agent can comfortably navigate.
Coverage is very strong for spreadsheet work: file discovery/creation, cell read/write/format, rows/columns, tables, charts, PivotTables, conditional formatting, data validation, import, PDF export, and code execution are all present, with connections and database querying added. Minor gaps remain, such as file-level rename/delete (files_delete only clears cells) and export to CSV/XLSX, but agents can work around these.
Publisher details
- Operator
- tabulya.ru · Publisher source
- Operator website
- https://tabulya.ru · Publisher source
- Vendor relationship
- First-party · Publisher source
- Documentation
- https://tabulya.ru/docs/ai/mcp-server · Publisher source
- Privacy policy
- Not provided
- Trust center
- Unknown
- Restrictions
- Unknown
Available Tools
68 toolsconnections_agent_getARead-onlyIdempotentInspect
Returns a single agent connection: its plan (endpoints the AI configured), API schema, secret binding names (never their values), and last test status. Signature: connections_agent_get(connection_uuid, space_uuid?).
Inputs: connection_uuid (required), space_uuid. Returns: name, plan, uuid, apiSchema, +10 more
connections_agent_listARead-onlyIdempotentInspect
Lists the agent connections (AI-configured REST API integrations) of a Tabulya space: uuid, name, and createdDate. Use connections_agent_get for a single connection's plan and API schema. Signature: connections_agent_list(space_uuid?).
Inputs: space_uuid. Returns: space_uuid, connections
connections_get_schemasARead-onlyIdempotentInspect
Returns full table schemas (tables + columns + types) for database connections in a Tabulya space. Run connections_list first to discover connection uuids; then call this with their connection_ids. If connection_ids is empty it returns schemas for every connection in the space — but if the space has none, the result tells you to call connections_list. For very large databases, prefer connections_search_tables (semantic search) over fetching every schema. Signature: connections_get_schemas(space_uuid?, connection_ids?[]).
Inputs: space_uuid, connection_ids. Returns: note, errors, schemas
connections_listARead-onlyIdempotentInspect
First step for working with databases. Lists the database connections (Postgres, MySQL, ClickHouse, etc.) configured for a Tabulya space. Returns each connection's uuid, name, type, and createdDate. Flow: connections_list → connections_get_schemas (or connections_search_tables for big databases) → use a connection uuid to query its tables. Signature: connections_list(space_uuid?).
Inputs: space_uuid. Returns: connections
connections_search_tablesARead-onlyIdempotentInspect
Semantic search over a single connection's tables — the scalable alternative to connections_get_schemas for large databases. Returns top-K tables ranked by relevance to the query (pgvector embeddings of table metadata). Get the connection_uuid from connections_list first. Signature: connections_search_tables(connection_uuid, query, space_uuid?, top_k?=20).
Inputs: query (required), connection_uuid (required), top_k, space_uuid. Returns: tables
files_add_chartAInspect
Adds a native (Excel-style) chart that visualizes data from an A1 cell range — no Python. Native charts stay live as the referenced cells change and are editable in the Chart Editor. The chart floats over the grid anchored at a cell; it does not occupy cells. data_range is a rectangular A1 range (default: first row = series headers, first column = category labels, other columns = series). When position is null the chart is auto-placed two columns right of the data; when a position is given and it would overlap another chart or non-empty cells, the chart is moved down to free space and the result reports the final position. Requires edit permission. Signature: files_add_chart(file_uuid, chart_type, data_range, sheet?, chart_name?, title?, position?, width_cells?, height_cells?, transpose?, first_row_headers?, first_col_labels?, x_column?).
Inputs: file_uuid (required), chart_type (required), data_range (required), sheet, +15 more. Returns: name, moved, chart_id, position, +2 more
files_add_tableADestructiveInspect
Creates a structured data table from a 2D array of values in one call (named, with headers). Data tables are discouraged by default — only use this when the user explicitly requests a data table or the file already uses data tables; otherwise write plain cells with files_write (or files_write_batch) and style the header row. cell: the table's top-left anchor. table_name: the table's name. values: a 2D array of strings (first row is the column headers when first_row_is_header is true, the default). Don't place over existing data (check with files_has_data). For plain non-table cells use files_write; to turn existing cells into a table use files_convert_to_table. Requires edit permission. Signature: files_add_table(file_uuid, cell, table_name, values, sheet?, first_row_is_header?).
Inputs: cell (required), values (required), file_uuid (required), table_name (required), +2 more. Returns: cell, file_uuid, sequence_number
files_bordersAInspect
Applies (or clears) cell borders on an A1 range. sides: which edges — all, inner, outer, horizontal, vertical, left, top, right, bottom, or clear (remove borders). line: line1 (thin, default), line2 (medium), line3 (thick), double, dotted, dashed. color: hex like '#000000' (default black). Requires edit permission; does not change values. Note: re-applying borders that exactly match the existing ones toggles them off. Signature: files_borders(file_uuid, range, sheet?, sides?, line?, color?).
Inputs: range (required), file_uuid (required), line, color, +2 more. Returns: note, range, changed, file_uuid, +1 more
files_conditional_formatsARead-onlyIdempotentInspect
Lists the conditional formatting rules defined in a Tabulya file, so you can inspect existing rules before adding or replacing them. Scans every sheet unless 'sheet' is given. Each rule is { sheet, id, range (A1), type ('formula', 'color_scale', 'data_bar' or 'aggregate'), apply_to_blank; for formula rules also formula (the condition) and style (applied bold/italic/underline/strikethrough/text_color/fill_color); for color scales also color_scale; for data bars also data_bar; aggregate rules come from Excel workbooks and are computed over the whole range (duplicate/unique values, top or bottom N, above or below average) — they carry rule and style, and cannot be created or edited here. Read-only; reflects the file's latest live state. Signature: files_conditional_formats(file_uuid, sheet?).
Inputs: file_uuid (required), sheet. Returns: count, file_uuid, conditional_formats
files_contextARead-onlyIdempotentInspect
Returns structured context about a Tabulya spreadsheet from its latest live state: per-sheet schema (sheets and their in-grid data tables with A1 ranges and column headers), native charts (id, name, type, anchor cell, data range — use the id with files_update_chart / files_delete_chart), PivotTables (name, bounds, source, field areas — use the name with files_set_pivot_table / files_delete_pivot_table), and the code-cell dependency graph (each code cell's position, language, a code preview, and the cells/ranges it reads). Use it to understand a file before reading or editing specific ranges. Signature: files_context(file_uuid).
Inputs: file_uuid (required). Returns: sheets, tablos, file_uuid, sequence_number
files_convert_to_tableAInspect
Converts a rectangular range of cells into a structured data table (like an Excel Table): named, with column headers, sortable. Only use when the user explicitly requests a data table or the file already uses data tables; otherwise keep the data as plain cells. range must be a single rectangle (e.g. 'A1:D20') and must NOT overlap an existing table or code-cell output. first_row_is_header (default true): use the first row as the column names; false generates 'Column 1', 'Column 2', … and keeps all rows as data. table_name: optional (auto-generated if omitted). Requires edit permission. Signature: files_convert_to_table(file_uuid, range, table_name?, first_row_is_header?, sheet?).
Inputs: range (required), file_uuid (required), sheet, table_name, +1 more. Returns: range, file_uuid, sequence_number
files_createAInspect
Creates a new, empty Tabulya spreadsheet file in a space and returns its uuid. Requires edit permission in the space (or view permission if 'private' is true). Use the returned file_uuid with files_write / files_run / sheets_* to populate it. Signature: files_create(name, space_uuid?, private?).
Inputs: name (required), private, space_uuid. Returns: name, file_uuid, space_uuid
files_deleteADestructiveIdempotentInspect
Deletes (clears) cells in an A1 range, INCLUDING code cells (Python/Formula), charts and data tables — a touched table is removed entirely. Requires edit permission. Use this to remove or fix content created by files_write/files_run (which cannot overwrite a table). Within the data rows of a server table the range clears their values through the tables service and the table stays; a range covering all of its data rows removes the table, as Delete does in the app. Writes synchronously via collab. Signature: files_delete(file_uuid, range, sheet?).
Inputs: range (required), file_uuid (required), sheet. Returns: note, range, cleared, file_uuid, +1 more
files_delete_chartADestructiveIdempotentInspect
Deletes a native chart by id. Chart ids come from files_add_chart or the sheet context. This deletes only native charts (created with files_add_chart or the Insert menu); Python (Altair) charts are code cells — remove them with files_delete. Requires edit permission. Signature: files_delete_chart(file_uuid, chart_id, sheet?).
Inputs: chart_id (required), file_uuid (required), sheet. Returns: file_uuid, sequence_number
files_delete_conditional_formatADestructiveInspect
Removes a conditional formatting rule by its id (obtained from files_conditional_formats). Requires edit permission; does not change cell values. Signature: files_delete_conditional_format(file_uuid, id, sheet?).
Inputs: id (required), file_uuid (required), sheet. Returns: file_uuid, sequence_number
files_delete_pivot_tableADestructiveIdempotentInspect
Deletes a PivotTable by name. PivotTable names come from files_context or files_set_pivot_table. Requires edit permission. Signature: files_delete_pivot_table(file_uuid, pivot_table_name).
Inputs: file_uuid (required), pivot_table_name (required). Returns: deleted, file_uuid, sequence_number
files_delete_rows_colsADestructiveInspect
Deletes entire rows or columns from a sheet, shifting remaining content up/left. axis: 'rows' or 'columns'. indices: which to delete — column letters (e.g. ['A','C']) for columns, or row numbers (e.g. [1,5,10]) for rows. NOTE: distinct from files_delete, which only CLEARS cell values without removing the row/column. Requires edit permission. Signature: files_delete_rows_cols(file_uuid, axis, indices[], sheet?).
Inputs: axis (required), indices (required), file_uuid (required), sheet. Returns: axis, deleted, file_uuid, sequence_number
files_delete_validationADestructiveInspect
Removes a data-validation rule (checkbox/dropdown/etc.) by its id (from files_validations). Requires edit permission; does not change cell values. Signature: files_delete_validation(file_uuid, id, sheet?).
Inputs: id (required), file_uuid (required), sheet. Returns: file_uuid, sequence_number
files_dependenciesARead-onlyIdempotentInspect
Traces the dependency graph around a cell. 'forward' returns what the code cell at the given cell READS (its input ranges, as 'reads'); 'reverse' returns the code cells that DEPEND ON the given cell (as 'read_by'); 'both' (default) returns both. Use to understand impact before editing a cell or to find a code cell's inputs. Read-only. Signature: files_dependencies(file_uuid, cell, sheet?, direction?).
Inputs: cell (required), file_uuid (required), sheet, direction. Returns: file_uuid, dependencies
files_export_pdfARead-onlyIdempotentInspect
Exports sheets as a PDF with Excel-style print semantics and returns it base64-encoded. Options mirror Excel's Page Setup: paper size, orientation, margins (inches), zoom or fit-to-pages scaling, page order, centering, gridlines and headings, header/footer with Excel codes (&P page, &N total, &D date, &T time, &F file, &A sheet, &B bold); numbered date and time codes (&D1…&D6, &T1, &T2) take their strings from printedValues, since locale and timezone are the caller's to decide. Each sheet may carry its own saved Page Setup (page setup, print area, print titles, manual breaks — set in the app or imported from xlsx); export uses it by default. sheetOptions[sheetId] overrides the saved values field by field (an empty string or empty list clears one for this export), sheetOptions[sheetId].pageSetup overrides the saved page setup, and the top-level pageSetup applies only to sheets without a saved one. repeatFrozenRows/repeatFrozenCols repeat whatever each sheet has frozen, so a book prints every sheet with its own header; explicit repeat rows/columns win over them. Images placed in cells are printed from the file itself (PNG and JPEG; other formats become a framed placeholder). Rasters a document may unpack are capped: past the cap an image is drawn as a framed placeholder instead. Charts print from the snapshot the client stores in the file, so they come out on every sheet, server-side too; chartImages (base64 PNG keyed by "sheetId!x,y") overrides it, and a chart without a snapshot becomes a framed placeholder — a workbook imported through this API carries no snapshots until a client opens it, so pass chartImages to print its charts. Read-only. Signature: files_export_pdf(file_uuid, options?).
Inputs: file_uuid (required), options. Returns: size, file_uuid, pdf_base64, content_type
files_flatten_tableADestructiveInspect
Flattens a data table back into plain cell values (the inverse of files_convert_to_table): the table's values are written into the grid and the table wrapper is removed; hidden columns go right of the shown ones, so those cells must be empty. cell: any cell inside the table (e.g. its top-left anchor 'A5'). Works on code-output tables too (bakes the computed values). Requires edit permission. Signature: files_flatten_table(file_uuid, cell, sheet?).
Inputs: cell (required), file_uuid (required), sheet. Returns: cell, file_uuid, sequence_number
files_formatAInspect
Applies cell formatting to an A1 range in a Tabulya spreadsheet (bold/italic/underline/strikethrough, text & fill color, horizontal & vertical alignment, wrapping, font size, number format like currency/percent/accounting/text, thousands separators, date/time display format, and custom_format — an ECMA-376 number format code for anything the discrete tokens cannot express: section colors, conditions, fractions, padding, literals). Codes are stored in the invariant (English) notation, the same one files_text_formats returns. Pass clear:true to remove ALL formatting from the range. Requires edit permission; non-destructive (does not change values). Pass only the properties you want to change. Applied through the collaboration server. Signature: files_format(file_uuid, range, sheet?, clear?, bold?, italic?, underline?, strikethrough?, text_color?, fill_color?, align?, vertical_align?, wrap?, number_format?, decimals?, currency_symbol?, negative?, numeric_commas?, custom_format?, date_time?, font_size?).
Inputs: range (required), file_uuid (required), bold, wrap, +17 more. Returns: range, file_uuid, sequence_number
files_get_infoARead-onlyIdempotentInspect
Returns metadata for a single Tabulya file: name, spaceId, createdDate, updatedDate, the latest saved sequenceNumber, and your canView/canEdit permissions. Use to inspect one file before reading or editing it (files_list discovers many; this targets one by uuid). Read-only. Signature: files_get_info(file_uuid).
Inputs: file_uuid (required). Returns: name, uuid, canEdit, canView, +4 more
files_has_dataARead-onlyIdempotentInspect
Checks whether an A1 range is already occupied — cell values, code/table output, or merged cells (each can cause a spill). Use before placing a table or code cell to avoid spill errors. Returns { has_data: true/false }. Read-only; reflects the file's latest live state. Signature: files_has_data(file_uuid, range, sheet?).
Inputs: range (required), file_uuid (required), sheet. Returns: range, has_data, file_uuid
files_importADestructiveInspect
Imports a CSV, Excel, or Parquet file into a Tabulya spreadsheet. file_name's extension picks the format: .csv (→ a data table at 'cell'), .xlsx/.xls/.xlsm/.xlsb/.ods (→ each workbook sheet becomes a new sheet; 'cell'/'sheet' ignored), .parquet (→ a data table at 'cell'). file_data is the file's bytes, base64-encoded. Requires edit permission. Signature: files_import(file_uuid, file_name, file_data, sheet?, cell?).
Inputs: file_data (required), file_name (required), file_uuid (required), cell, +1 more. Returns: summary, file_uuid, sequence_number
files_insertADestructiveInspect
Inserts blank rows or columns into a sheet, shifting existing content down/right. axis: 'rows' or 'columns'. at: the reference line — a column letter (e.g. 'C') for columns, or a row number (e.g. '5') for rows. side: 'before' (default) inserts on the low side (left of the column / above the row), so content at 'at' shifts; 'after' inserts on the high side (right/below). Either way the new line inherits the formatting of the 'at' line. count: how many to insert (default 1). Tables and formulas adjust automatically (recomputed on the collaboration server). Requires edit permission. Signature: files_insert(file_uuid, axis, at, count?, side?, sheet?).
Inputs: at (required), axis (required), file_uuid (required), side, +2 more. Returns: axis, inserted, file_uuid, sequence_number
files_listARead-onlyIdempotentInspect
Lists Tabulya spreadsheets in a space, most-recently-updated first. Returns each file's uuid, name, createdDate and updatedDate; nextCursor is a string when more files remain (pass it back as cursor to get the next page) and null on the last page. Use this to discover files before reading them with files_read. Signature: files_list(space_uuid?, limit?=100, cursor?).
Inputs: limit, cursor, space_uuid. Returns: files, nextCursor
files_mergeADestructiveInspect
Merges a range of cells into one (keeping only the top-left value, clearing the rest), or splits merged cells back when unmerge=true. Use by_rows=true to merge each row of the range separately (one merged cell per row), and center=true to also center the merged content horizontally. Requires edit permission. Applied through the collaboration server. Signature: files_merge(file_uuid, range, sheet?, unmerge?, by_rows?, center?).
Inputs: range (required), file_uuid (required), sheet, center, +2 more. Returns: note, range, merged, file_uuid, +1 more
files_moveADestructiveInspect
Moves a rectangular block of cells (values, formats, code) to a new location on the same sheet. source: an A1 rectangle (e.g. 'A1:C10'). target: the top-left cell where the block should land (e.g. 'F1'). References from code/formulas adjust as on the collaboration server. Requires edit permission. Signature: files_move(file_uuid, source, target, sheet?).
Inputs: source (required), target (required), file_uuid (required), sheet. Returns: file_uuid, sequence_number
files_move_rows_colsADestructiveInspect
Reorders whole columns or rows: cuts a column/row range and reinserts it at a destination index, shifting the others and adjusting formula/table references (distinct from files_move, which relocates a rectangular block of cells). axis='columns': range and to are column letters (e.g. range 'B:D', to 'H'); axis='rows': range and to are row numbers (e.g. range '2:5', to '10'). 'to' is the index the range lands before. Requires edit permission. Signature: files_move_rows_cols(file_uuid, axis, range, to, sheet?).
Inputs: to (required), axis (required), range (required), file_uuid (required), +1 more. Returns: file_uuid, sequence_number
files_readARead-onlyIdempotentInspect
Reads cell values from a Tabulya spreadsheet file. Pass either 'range' (a single A1-notation range) or 'ranges' (an array of A1 ranges, read in one shot). Returns 2D arrays of display values (strings) from the file's latest live state. If a code cell in a range failed to run, its display value is empty and the failure is reported in a 'code_errors' array ([{cell, error, std_err}]). Signature: files_read(file_uuid, range? | ranges?[], sheet?).
Inputs: file_uuid (required), range, sheet, ranges. Returns: range, ranges, values, file_uuid, +2 more
files_read_codeARead-onlyIdempotentInspect
Returns the SOURCE CODE of a code cell (Python, SQL connection, or Formula) at a given cell, plus its run stdout/stderr. files_read returns only the computed DISPLAY values; use this to see (and then modify) the actual code. Pass any cell within the code cell's output — the anchor is resolved automatically. Returns code_cell: null when there is no code cell there. Signature: files_read_code(file_uuid, cell, sheet?).
Inputs: cell (required), file_uuid (required), sheet. Returns: code_cell, file_uuid
files_replaceADestructiveInspect
Finds and replaces text across cells in one transaction (the write counterpart of files_search). Only TEXT cells are changed — numbers, formulas, and code are left untouched. case_sensitive (default false) and whole_cell (default false: substring match; true: only cells equal to find). sheet scopes the replace to one sheet (defaults to all sheets). Returns { replaced: , skipped: <matches left as-is: numbers, formula output, table cells> }. Requires edit permission. Signature: files_replace(file_uuid, find, replace, sheet?, case_sensitive?, whole_cell?).
Inputs: find (required), replace (required), file_uuid (required), sheet, +2 more. Returns: note, skipped, replaced, file_uuid, +1 more
files_rerunAInspect
Re-runs existing code cells (Python/SQL) WITHOUT changing their code — e.g. to refresh external data. range: an A1 selection of code cells to re-run; omit it to re-run every code cell on the sheet. Execution is ASYNCHRONOUS: returns a task id immediately; read results with files_read after a few seconds. (To set new code, use files_run; this only re-runs unchanged code.) Requires edit permission. Signature: files_rerun(file_uuid, range?, sheet?).
Inputs: file_uuid (required), range, sheet. Returns: note, status, task_id, file_uuid
files_resizeAIdempotentInspect
Sets column widths or row heights, in pixels. axis: 'columns' or 'rows'. size: pixel size (1-2000). indices: which to resize — column letters (e.g. ['A','C']) for columns, or row numbers (e.g. [1,5]) for rows. Omit indices to set the DEFAULT size for ALL columns/rows on the sheet. ('auto' fit-to-content is not available headless.) Requires edit permission. Signature: files_resize(file_uuid, axis, size, indices?, sheet?).
Inputs: axis (required), size (required), file_uuid (required), sheet, +1 more. Returns: file_uuid, sequence_number
files_runADestructiveInspect
Sets a code cell at the given position and runs it in a sandboxed cloud worker, writing its result back into the spreadsheet. Prefer files_set_formula whenever a formula can accomplish the task; only use code when the functionality is not available in formulas (charts, machine learning, complex transforms, web/API requests). Requires edit permission. language='python' (pandas/numpy) runs code; language='connection' runs an SQL query against a database connection (pass connection_uuid from connections_list; 'code' is the SQL) and lands the result table at the cell. Execution is ASYNCHRONOUS: returns a task id immediately; read the cell with files_read after a few seconds. Signature: files_run(file_uuid, cell, language, code, sheet?, connection_uuid?).
Inputs: cell (required), code (required), language (required), file_uuid (required), +2 more. Returns: cell, note, status, task_id, +1 more
files_run_cancelAIdempotentInspect
Cancels a previously started files_run task by its task_id (only the task's owner can cancel). Marks the task so the worker won't pick it up; a task that is already running may still finish. Returns whether it was cancelled. Signature: files_run_cancel(task_id, file_uuid?).
Inputs: task_id (required), file_uuid. Returns: fileUuid, taskUuid, cancelled, previousStatus
files_run_statusARead-onlyIdempotentInspect
Returns the execution status of a files_run task by its task_id: 'queued' (accepted, not yet picked up by a worker), 'running', 'succeeded' (the worker executed the cell), or 'failed' (the worker could not run the task — see 'error'). NOTE: 'succeeded' means the worker ran the code; a code-level error (e.g. a Python exception) still reports 'succeeded' here — read the cell with files_read and inspect its 'code_errors' to detect code-level failures. Signature: files_run_status(task_id, file_uuid?).
Inputs: task_id (required), file_uuid. Returns: error, state, task_id
files_searchARead-onlyIdempotentInspect
Searches cell values across a Tabulya spreadsheet file and returns the matching cells, so you can locate data without reading whole sheets. Searches every sheet unless 'sheet' is given. Each hit is { sheet, cell (A1), x, y (1-based coords), value }. Read-only; searches the file's latest live state. Options: case_sensitive; whole_cell (match only cells whose entire value equals the query); search_code (also match the source of code cells, not just their displayed output). Signature: files_search(file_uuid, query, sheet?, case_sensitive?, whole_cell?, search_code?).
Inputs: query (required), file_uuid (required), sheet, whole_cell, +2 more. Returns: count, results, file_uuid, truncated, +1 more
files_set_checkboxAInspect
Adds checkboxes to a range (a boolean data-validation rendered as a checkbox in each cell; the cell value is TRUE/FALSE). Pass 'id' (from files_validations) to replace that rule; omit to create. Requires edit permission; does not change existing values. Signature: files_set_checkbox(file_uuid, range, sheet?, id?).
Inputs: range (required), file_uuid (required), id, sheet. Returns: range, file_uuid, sequence_number
files_set_color_scaleAInspect
Adds (or replaces) a color-scale conditional formatting rule on a range: cells are filled with a gradient computed from their numeric value (a heatmap). Provide 'thresholds' (2+, ordered low→high), each { value_type: 'min'|'max'|'number'|'percent'|'percentile', value (number; required for number/percent/percentile, 0-100 for percent/percentile; omit for min/max), color: '#RRGGBB' }. Colors interpolate between thresholds. Examples: red→green is [{value_type:'min',color:'#FF0000'},{value_type:'max',color:'#00FF00'}]; a traffic light adds {value_type:'percentile',value:50,color:'#FFFF00'} in the middle. Pass 'id' (from files_conditional_formats) to replace that rule; omit to create. Optional auto_contrast_text flips text black/white for readability. Requires edit permission; does not change values. Signature: files_set_color_scale(file_uuid, range, thresholds, sheet?, id?, auto_contrast_text?).
Inputs: range (required), file_uuid (required), thresholds (required), id, +2 more. Returns: range, file_uuid, sequence_number
files_set_conditional_formatAInspect
Adds (or replaces) a formula-based conditional formatting rule on a range: cells where the formula evaluates truthy get the given style. The formula is a spreadsheet formula referencing cells relative to the top-left of the range (like Excel conditional formatting), e.g. 'A1>100', 'ISBLANK(A1)', 'A1="done"'. At least one style property must be set. Pass 'id' (from files_conditional_formats) to replace that existing rule instead of creating a new one — omit it to create a new rule. Requires edit permission; does not change values. Signature: files_set_conditional_format(file_uuid, range, formula, sheet?, id?, bold?, italic?, underline?, strikethrough?, text_color?, fill_color?, apply_to_blank?).
Inputs: range (required), formula (required), file_uuid (required), id, +8 more. Returns: range, file_uuid, sequence_number
files_set_custom_validationAInspect
Adds a custom formula validation to a range — Excel's "Custom" data-validation type: the value is accepted when the formula evaluates to TRUE. Write the formula for the top-left cell of the range; relative references shift for every other cell, as in Excel. The formula sees the value being entered, so "=A1>0" on A1:A10 rejects a negative number as it is typed. strict (default true): reject invalid entries; false warns. Pass 'id' (from files_validations) to replace. Requires edit permission. Signature: files_set_custom_validation(file_uuid, range, formula, sheet?, id?, strict?, ignore_blank?).
Inputs: range (required), formula (required), file_uuid (required), id, +3 more. Returns: range, file_uuid, sequence_number
files_set_datetime_validationAInspect
Adds a date/time data-validation to a range. Constraints (the value passes if it satisfies at least one): date_min/date_max (inclusive date bounds; both together = 'between'), date_equals/date_not_equals (date lists), date_after/date_before (STRICT exclusive bounds; both together express 'not between'), time_min/time_max (inclusive time bounds), time_equals/time_not_equals (time lists). Dates are 'YYYY-MM-DD', times 'HH:MM' or 'HH:MM:SS'; date-time cell values are compared to date constraints by day. Flags: require_date/require_time (the value must include a date/time; set automatically when date/time constraints are given), prohibit_date/prohibit_time (must NOT include one). At least one flag or constraint required. strict (default true): reject invalid entries; false warns. Pass 'id' (from files_validations) to replace. Requires edit permission. Signature: files_set_datetime_validation(file_uuid, range, sheet?, id?, strict?, require_date?, require_time?, prohibit_date?, prohibit_time?, date_min?, date_max?, date_equals?, date_not_equals?, date_after?, date_before?, time_min?, time_max?, time_equals?, time_not_equals?).
Inputs: range (required), file_uuid (required), id, sheet, +15 more. Returns: range, file_uuid, sequence_number
files_set_dropdownAInspect
Adds a dropdown (list data-validation) to a range: each cell gets a dropdown of allowed values. Provide either 'values' (an explicit list of strings) or 'source_range' (an A1 range whose cell values are the options) — exactly one. strict (default true): reject entries not in the list; false: warn but allow. Pass 'id' (from files_validations) to replace that rule; omit to create. Requires edit permission; does not change existing values. Signature: files_set_dropdown(file_uuid, range, values?|source_range?, sheet?, id?, strict?).
Inputs: range (required), file_uuid (required), id, sheet, +3 more. Returns: range, file_uuid, sequence_number
files_set_formulaADestructiveInspect
Sets a spreadsheet FORMULA at the given cell and returns once it is computed (formulas evaluate synchronously — unlike files_run code, no waiting). This is the preferred way to compute values — use a formula whenever it can do the task; fall back to files_run code only when the functionality is not available in formulas. Use A1-style formulas like 'SUM(A1:A10)', 'B2*1.2', 'VLOOKUP(...)'; a leading '=' is optional. Excel's English notation (',' between arguments, '.' as the decimal point) is converted to the Russian notation the file stores and files_read_code returns (';' between arguments, ',' as the decimal point); formulas already in Russian notation are kept as is. For a tiny in-cell chart use SPARKLINE, e.g. 'SPARKLINE(B2:M2, {"charttype", "column"})'. Requires edit permission; applied through the collaboration server. To read the computed value, files_read the cell. Signature: files_set_formula(file_uuid, cell, formula, sheet?).
Inputs: cell (required), formula (required), file_uuid (required), sheet. Returns: cell, note, file_uuid, sequence_number
files_set_number_validationAInspect
Adds a numeric data-validation to a range: the number must satisfy at least one constraint. Provide any of: min and/or max (inclusive bounds — a value passes if min<=v<=max), greater_than/less_than (STRICT exclusive bounds; both together express 'not between': valid when v<less_than OR v>greater_than), equals (list — value must equal one of these), not_equals (list — value must differ from all of these). At least one constraint required. require_integer (default false): only whole numbers pass, a fractional value fails regardless of the bounds (Excel's "whole" rule). strict (default true): reject invalid entries; false warns. Pass 'id' (from files_validations) to replace. Requires edit permission. Signature: files_set_number_validation(file_uuid, range, sheet?, id?, strict?, min?, max?, greater_than?, less_than?, equals?, not_equals?, require_integer?).
Inputs: range (required), file_uuid (required), id, max, +8 more. Returns: range, file_uuid, sequence_number
files_set_pivot_tableAInspect
Creates or reconfigures a PivotTable: a live cross-tabulation that groups source rows by fields and aggregates values per group; it recomputes when the source changes and exports to Excel as a real PivotTable. action='create' needs source (an A1 range whose first row is headers, e.g. 'A1:E500', or a table name) and at least one values entry; destination is 'new_sheet' (default) or a top-left cell like 'F1'. action='update' needs pivot_table_name (from files_context) and applies a merge: null leaves an area unchanged, an empty array clears it. Reference source columns by header name. rows/columns entries: {field, label?, sort? (ascending|descending|source_order), show_totals?, group_by? (seconds|minutes|hours|days|day_of_week|months|quarters|years), numeric_interval?}. values entries: {field, aggregation? (sum|count|count_numbers|distinct_count|average|median|max|min|product|stddev|stddevp|var|varp), name?, show_as? (percent_of_row_total|percent_of_column_total|percent_of_grand_total|running_total_in_rows|...|index)}. filters entries: {field, include?, exclude?}. Requires edit permission. Signature: files_set_pivot_table(file_uuid, action, pivot_table_name?, sheet?, source?, destination?, rows?, columns?, values?, filters?, layout?, values_layout?, row_grand_total?, column_grand_total?, subtotal_position?).
Inputs: action (required), file_uuid (required), rows, sheet, +11 more. Returns: created, updated, file_uuid, pivot_table, +2 more
files_set_table_filterAInspect
Filters a data table's rows with its native column filters (Excel AutoFilter); only display is affected, the data is unchanged. cell: any cell inside the table. column: 0-based column index within the table (omit for action 'show_all' to clear every column). action: 'show_only' shows only rows whose value matches one of values; 'hide' hides them in addition to already hidden ones; 'show_all' clears the column's filter (or all filters without column); 'condition' applies a comparison (numeric or date columns) from condition: { op: equals|not_equals|greater_than|greater_than_or_equal|less_than|less_than_or_equal, value, op2?, value2?, connector?: and|or }. Values are matched as plain unformatted text; for blank cells pass '(Пустые)' (or '' / '(Blanks)'). Requires edit permission. Signature: files_set_table_filter(file_uuid, cell, action, column?, values?, condition?, sheet?).
Inputs: cell (required), action (required), file_uuid (required), sheet, +3 more. Returns: file_uuid, sequence_number
files_set_text_validationAInspect
Adds a text data-validation to a range: cell text must satisfy the given constraints. Provide any of: min_length/max_length (inclusive character-count bounds), exactly (list — the value must equal one of these), contains (list — must contain one of these), not_contains (list — must contain none of these). Matching is case-insensitive. At least one constraint required. strict (default true): reject invalid entries; false warns. Pass 'id' (from files_validations) to replace. Requires edit permission. Signature: files_set_text_validation(file_uuid, range, sheet?, id?, strict?, min_length?, max_length?, exactly?, contains?, not_contains?).
Inputs: range (required), file_uuid (required), id, sheet, +6 more. Returns: range, file_uuid, sequence_number
files_table_columnsAInspect
Renames or shows/hides the columns of a data table. cell: any cell inside the table. columns: a list of changes, each { name (the column's CURRENT name), new_name? (rename to this), hidden? (true hides the column, false shows it) }. Only the columns you list are changed; the rest stay the same. Requires edit permission; does not change the table's data. Signature: files_table_columns(file_uuid, cell, columns[], sheet?).
Inputs: cell (required), columns (required), file_uuid (required), sheet. Returns: cell, file_uuid, sequence_number
files_table_deleteADestructiveInspect
Deletes columns or rows from INSIDE an existing data table (distinct from files_delete_rows_cols, which removes sheet-level rows/columns). cell: any cell inside the table. axis='columns' or 'rows'. index: 0-based position within the table. count: how many to remove (default 1). Server tables (rows on the tables service): rows are counted in the table's current sort and filter (at most 1,000,000 per call), and columns as files_context lists them. Requires edit permission. Signature: files_table_delete(file_uuid, cell, axis, index, count?, sheet?).
Inputs: axis (required), cell (required), index (required), file_uuid (required), +2 more. Returns: file_uuid, sequence_number
files_table_insertAInspect
Inserts blank columns or rows INSIDE an existing data table (distinct from files_insert, which inserts sheet-level rows/columns). cell: any cell inside the table. axis='columns' or 'rows'. index: 0-based position within the table to insert at. count: how many to insert (default 1). Server tables (rows on the tables service): rows are inserted only when the table has no sort or filter; new columns always go after its data columns (index is ignored), named by names and typed by kind. Requires edit permission. Signature: files_table_insert(file_uuid, cell, axis, index, count?, names?, kind?, sheet?).
Inputs: axis (required), cell (required), index (required), file_uuid (required), +4 more. Returns: file_uuid, sequence_number
files_table_metaAInspect
Changes a data table's metadata (without touching its data): rename it, toggle its name bar / column-header row / alternating row colors, or set whether the first data row is the header row. cell: any cell inside the table. Pass only the properties to change. Requires edit permission. Signature: files_table_meta(file_uuid, cell, name?, show_name?, show_columns?, alternating_colors?, first_row_as_header?, sheet?).
Inputs: cell (required), file_uuid (required), name, sheet, +4 more. Returns: cell, file_uuid, sequence_number
files_table_sortAInspect
Sorts a data table by one or more of its columns. cell: any cell inside the table (e.g. its anchor 'A1'). sort: an array of { column, direction } applied in order, where column is the 0-based column index within the table and direction is 'asc' or 'desc'. Pass an empty array to clear the sort and restore the original row order. Requires edit permission. Signature: files_table_sort(file_uuid, cell, sort, sheet?).
Inputs: cell (required), sort (required), file_uuid (required), sheet. Returns: file_uuid, sequence_number
files_text_formatsARead-onlyIdempotentInspect
Reads the cell formatting in an A1 range (the inverse of files_format), so you can inspect existing styles before changing them. Returns one entry per cell that has a non-default format: { cell (A1), and the set properties among bold/italic/underline/strikethrough/text_color/fill_color/align/vertical_align/wrap/number_format/negative/decimals/currency_symbol/numeric_commas/custom_format/date_time/font_size } — the same vocabulary files_format accepts, so a read format can be re-applied. custom_format holds an ECMA-376 number format code in the invariant (English) notation and takes precedence over the discrete tokens when both are present. Read-only; reflects the file's latest live state. Signature: files_text_formats(file_uuid, range, sheet?).
Inputs: range (required), file_uuid (required), sheet. Returns: count, formats, file_uuid, truncated
files_update_chartAInspect
Updates an existing native chart by id. Only the provided (non-null) fields change; everything else is preserved. To MOVE the chart set position; to RESIZE set width_cells/height_cells. Set chart_name to rename (display name, not the title); set title to '' to remove the title. series_columns replaces the plotted series (0-based column indexes within the data range, category column excluded). Requires edit permission. Signature: files_update_chart(file_uuid, chart_id, sheet?, chart_type?, data_range?, chart_name?, title?, position?, width_cells?, height_cells?, transpose?, first_row_headers?, first_col_labels?, x_column?, series_columns?).
Inputs: chart_id (required), file_uuid (required), sheet, title, +11 more. Returns: chart_id, file_uuid, sequence_number
files_validationsARead-onlyIdempotentInspect
Lists the data-validation rules in a Tabulya file, so you can inspect them before adding, replacing, or deleting. Scans every sheet unless 'sheet' is given. Each rule is { sheet, id, range (A1), kind ('checkbox'|'logical'|'dropdown'|'list'|'text'|'number'|'datetime'|'none') } plus its constraints under the same field names the files_set_* tools accept (values/source_range, min_length/max_length/exactly/contains/not_contains, min/max/greater_than/less_than/equals/not_equals, require_*/prohibit_* flags, date_*/time_* constraints) and 'strict'. Custom help/error texts appear as message_title/message_text/error_title/error_message; replacing by id via files_set_* rewrites the constraints but preserves these texts. Replacing by id overwrites all constraints, so read the current ones first. Extra markers: composite:true — the rule holds more conditions than these fields can express (do NOT blind-replace it); case_sensitive:true — a text rule matches case-sensitively (files_set_text_validation always writes case-insensitive rules); ignore_blank:false — the rule also validates empty cells (set tools always write true). Read-only; reflects the file's latest live state. Signature: files_validations(file_uuid, sheet?).
Inputs: file_uuid (required), sheet. Returns: count, file_uuid, validations
files_writeADestructiveInspect
Writes a rectangular block of LITERAL cell values into a Tabulya spreadsheet file, anchored at the given top-left cell. Preferred over files_add_table for tabular data (data tables are opt-in); when writing tabular data with a header row, use files_write_batch instead so the header gets styled (bold + fill_color) in the same atomic operation. Requires edit permission. Values are written as-is (formulas starting with '=' are rejected; so are writes overlapping or adjacent to data tables, except writes within the data rows of a server table: those go to the tables service — one table per call, values matching the column types). The write is applied through the collaboration server (visible to live editors and persisted). Signature: files_write(file_uuid, cell, values[][], sheet?).
Inputs: cell (required), values (required), file_uuid (required), sheet. Returns: cell, file_uuid, cols_written, rows_written, +1 more
files_write_batchADestructiveInspect
Applies MULTIPLE value-writes AND/OR format-changes to a Tabulya file in ONE call: the file is loaded once and everything ships as a SINGLE atomic collaboration transaction — either all of it applies, or (on any error) none does. Far cheaper than N separate calls, and lets you write data and style it atomically (writes apply first, then formats). Provide 'writes' and/or 'formats' (at least one). Each write: { cell (A1 anchor), values[][] (2D literals), sheet? } (same rules as files_write: literals only; no writes over/adjacent to data tables, except writes within the data rows of a server table: those go to the tables service, each write within one table; copies of a table sorted or filtered differently go in separate calls). Each format: { range (A1), sheet?, and any of bold/italic/underline/strikethrough/text_color/fill_color/align/wrap/number_format/custom_format/decimals/currency_symbol } (same as files_format). A top-level 'sheet' is the default for entries that omit one. Requires edit permission. Signature: files_write_batch(file_uuid, writes?[], formats?[], sheet?).
Inputs: file_uuid (required), sheet, writes, formats. Returns: file_uuid, writes_applied, formats_applied, sequence_number
sheets_colorAInspect
Sets (or clears) the tab color of a sheet. color: a CSS color string (e.g. '#1F77B4' or 'red'); pass an empty string to clear the color. Requires edit permission. Signature: sheets_color(file_uuid, sheet, color).
Inputs: color (required), sheet (required), file_uuid (required). Returns: file_uuid, sequence_number
sheets_createAInspect
Adds a new (empty) sheet to a Tabulya spreadsheet file. Requires edit permission. If 'name' is omitted, an auto-numbered name is chosen. Optionally insert the sheet before an existing one with 'insert_before'. Applied through the collaboration server. Signature: sheets_create(file_uuid, name?, insert_before?).
Inputs: file_uuid (required), name, insert_before. Returns: file_uuid, sheet_name, sequence_number
sheets_deleteADestructiveInspect
Deletes a sheet (and all its data) from a Tabulya spreadsheet file. Requires edit permission. The last remaining sheet cannot be deleted. This is destructive. Applied through the collaboration server. Signature: sheets_delete(file_uuid, sheet).
Inputs: sheet (required), file_uuid (required). Returns: file_uuid, sequence_number
sheets_duplicateAInspect
Duplicates a sheet (its cells, formats, code, tables). new_name: optional name for the copy (defaults to ' (копия)'); must be unique. Requires edit permission. Signature: sheets_duplicate(file_uuid, sheet, new_name?).
Inputs: sheet (required), file_uuid (required), new_name. Returns: file_uuid, sequence_number
sheets_listARead-onlyIdempotentInspect
Lists the sheets of a Tabulya spreadsheet file in tab order: name, position, tab color and frozen rows/columns. Use it to discover sheet names before reading or editing ranges (files_context also returns per-sheet tables and code cells, this one is just the tabs). Read-only. Signature: sheets_list(file_uuid).
Inputs: file_uuid (required). Returns: sheets, file_uuid, sequence_number
sheets_moveAInspect
Reorders a sheet. before: the name of the sheet to place it immediately before; omit to move it to the end. Requires edit permission. Signature: sheets_move(file_uuid, sheet, before?).
Inputs: sheet (required), file_uuid (required), before. Returns: file_uuid, sequence_number
sheets_renameAInspect
Renames an existing sheet in a Tabulya spreadsheet file. Requires edit permission. The new name must be unique within the file. Applied through the collaboration server. Signature: sheets_rename(file_uuid, sheet, new_name).
Inputs: sheet (required), new_name (required), file_uuid (required). Returns: file_uuid, sequence_number
sheets_set_frozenAIdempotentInspect
Freezes (pins) the top rows and/or left columns of a sheet so they stay visible while the rest of the sheet scrolls. frozen_rows / frozen_columns: how many to pin from the top / left (0 = none); pass 0 for both to unfreeze. Requires edit permission. Signature: sheets_set_frozen(file_uuid, sheet, frozen_rows, frozen_columns).
Inputs: sheet (required), file_uuid (required), frozen_rows (required), frozen_columns (required). Returns: file_uuid, sequence_number
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
AI-native team workspace — tables, documents, workflow automation, live dashboards & analytics
AI access to Quadratic spreadsheets: open files, run Python/SQL, query connected databases.
AI workspace for you, your team, and every agent. Tables, docs (images, 4K video), formulas.
Teable is an AI spreadsheet for business that combines a familiar spreadsheet interface with a structured database. Its remote MCP server lets AI agents inspect project and table structure, read and update records, and access the Teable API through OAuth with existing project permissions. Website: [teable.ai](https://teable.ai)
Related MCP Servers
AlicenseNot gradedqualityDmaintenanceConnects AI agents like Claude, Cursor, and Windsurf to Retable for seamless AI-assisted data management. Enables AI agents to manage and interact with Retable workspaces, tables, and data through natural language.12 npmMIT- FlicenseNot gradedqualityDmaintenanceEnables AI assistants to present interactive spreadsheets within conversations, allowing users to view, edit, import, and export Excel/CSV data directly through the chat interface.-
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to query tabular data (CSV/TSV) using natural language with cell-level citations, supporting multi-tenant workspaces, access control, and semantic search.27MIT
- AlicenseBqualityCmaintenanceConnects AI tools directly to Airtable, allowing users to query, create, update, and delete records using natural language.11148 npm87MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.