Tool tree
A local-first document utility workbench for people and browser agents.
Implementation
gradeNot graded
gradeNot graded
workspace_fileAction
List, select, inspect history, undo, or discover file-specific capabilities in the browser-local workspace. Use list before select and capabilities before complex work.
View tool JSON
{
"name": "workspace_file",
"kind": "act",
"impl": "imperative",
"description": "List, select, inspect history, undo, or discover file-specific capabilities in the browser-local workspace. Use list before select and capabilities before complex work.",
"inputSchema": {
"type": "object",
"properties": {
"action": {
"type": "string",
"enum": [
"list",
"select",
"history",
"undo",
"rename",
"capabilities",
"clear"
]
},
"fileId": {
"type": "string",
"description": "File ID returned by action=list; required for select and optional for history or capabilities."
},
"name": {
"type": "string",
"description": "New file name for action=rename."
},
"category": {
"type": "string",
"description": "Optional capability category returned by a previous capabilities call."
}
},
"required": [
"action"
],
"additionalProperties": false
}
}inspect_fileAction
Inspect a file locally by fileId, or use the selected file when omitted. Returns structural facts, a safe text preview, and compatibility warnings.
View tool JSON
{
"name": "inspect_file",
"kind": "act",
"impl": "imperative",
"description": "Inspect a file locally by fileId, or use the selected file when omitted. Returns structural facts, a safe text preview, and compatibility warnings.",
"inputSchema": {
"type": "object",
"properties": {
"fileId": {
"type": "string",
"description": "Optional file ID from workspace_file(list); omitted uses the selected file."
}
},
"additionalProperties": false
}
}extract_contentAction
Extract semantic content by fileId, or run a local DuckDB DataFrame sample/profile/query without creating a new version.
View tool JSON
{
"name": "extract_content",
"kind": "act",
"impl": "imperative",
"description": "Extract semantic content by fileId, or run a local DuckDB DataFrame sample/profile/query without creating a new version.",
"inputSchema": {
"type": "object",
"properties": {
"fileId": {
"type": "string",
"description": "Optional file ID from workspace_file(list); omitted uses the selected file."
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 100,
"description": "Maximum semantic rows, pages, or slides; defaults to 20."
},
"mode": {
"type": "string",
"enum": [
"semantic",
"sample",
"profile",
"query"
],
"description": "DataFrame modes apply only to table files."
},
"data": {
"type": "object",
"description": "Declarative DataFrame pipeline. Omit fields you do not need.",
"properties": {
"sheet": {
"type": "string",
"description": "Optional XLSX worksheet name for the selected dataset."
},
"joins": {
"type": "array",
"description": "Join other imported datasets by file ID. Use base.column and alias.column in select/filter/groupBy when needed.",
"items": {
"type": "object",
"properties": {
"fileId": {
"type": "string",
"description": "Dataset ID returned by workspace_file(list)."
},
"alias": {
"type": "string",
"description": "Optional source alias; defaults to joined_1, joined_2, and so on."
},
"sheet": {
"type": "string",
"description": "Optional XLSX worksheet name for this source."
},
"type": {
"type": "string",
"enum": [
"inner",
"left",
"right",
"full"
]
},
"leftColumn": {
"type": "string",
"description": "Column in the selected base dataset."
},
"rightColumn": {
"type": "string",
"description": "Column in the joined dataset."
}
},
"required": [
"fileId",
"leftColumn",
"rightColumn"
],
"additionalProperties": false
}
},
"unions": {
"type": "array",
"description": "Append rows from compatible imported datasets by column name.",
"items": {
"type": "object",
"properties": {
"fileId": {
"type": "string",
"description": "Dataset ID returned by workspace_file(list)."
},
"sheet": {
"type": "string",
"description": "Optional XLSX worksheet name for this source."
}
},
"required": [
"fileId"
],
"additionalProperties": false
}
},
"select": {
"type": "array",
"items": {
"type": "string"
},
"description": "Columns to project; qualified aliases are supported for joins."
},
"filters": {
"type": "array",
"items": {
"type": "object",
"properties": {
"column": {
"type": "string"
},
"operator": {
"type": "string",
"enum": [
"eq",
"neq",
"gt",
"gte",
"lt",
"lte",
"contains",
"starts_with",
"ends_with",
"in",
"is_null",
"not_null"
]
},
"value": {
"description": "Scalar value, or an array for the in operator."
}
},
"required": [
"column",
"operator"
],
"additionalProperties": false
}
},
"groupBy": {
"type": "array",
"items": {
"type": "string"
},
"description": "Grouping columns."
},
"aggregations": {
"type": "array",
"items": {
"type": "object",
"properties": {
"operation": {
"type": "string",
"enum": [
"count",
"sum",
"mean",
"min",
"max",
"median",
"nunique"
]
},
"column": {
"type": "string",
"description": "Omit only for count(*)."
},
"as": {
"type": "string",
"description": "Optional output column name."
}
},
"required": [
"operation"
],
"additionalProperties": false
}
},
"sort": {
"type": "array",
"items": {
"type": "object",
"properties": {
"column": {
"type": "string"
},
"direction": {
"type": "string",
"enum": [
"asc",
"desc"
]
}
},
"required": [
"column"
],
"additionalProperties": false
}
},
"offset": {
"type": "integer",
"minimum": 0
},
"limit": {
"type": "integer",
"minimum": 0
}
},
"additionalProperties": false
}
},
"additionalProperties": false
}
}patch_documentAction
Patch DOCX, PDF, or XLSX content as a new version using format-aware libraries. Pass fileId directly, batch related edits, and use validate:true when verification is required.
View tool JSON
{
"name": "patch_document",
"kind": "act",
"impl": "imperative",
"description": "Patch DOCX, PDF, or XLSX content as a new version using format-aware libraries. Pass fileId directly, batch related edits, and use validate:true when verification is required.",
"inputSchema": {
"type": "object",
"properties": {
"fileId": {
"type": "string",
"description": "Optional file ID from workspace_file(list); omitted uses the selected file."
},
"action": {
"type": "string",
"enum": [
"replace",
"redact",
"fill_template",
"set_table_cell",
"set_table_cells",
"append",
"prepend",
"insert_paragraphs",
"delete_paragraphs",
"set_paragraph_style",
"remove_comments",
"set_properties",
"rotate_pages",
"delete_pages",
"reorder_pages",
"watermark",
"fill_form",
"set_cells",
"create_workbook"
]
},
"find": {
"type": "string",
"minLength": 1,
"description": "Exact text for replace or redact."
},
"replacement": {
"type": "string",
"description": "Replacement text for replace."
},
"text": {
"type": "string",
"description": "Text for append, prepend, watermark, or DOCX table cell content."
},
"scope": {
"type": "object",
"description": "Optional DOCX replace/fill scope; choose a heading or 1-based paragraph.",
"properties": {
"heading": {
"type": "string"
},
"paragraph": {
"type": "integer",
"minimum": 1
}
},
"additionalProperties": false
},
"paragraph": {
"type": "integer",
"minimum": 1,
"description": "1-based DOCX paragraph target for structural operations."
},
"anchor": {
"type": "string",
"description": "Exact DOCX paragraph text used as a structural anchor."
},
"position": {
"type": "string",
"enum": [
"before",
"after"
],
"description": "Insert relative to paragraph or anchor; defaults to after."
},
"style": {
"type": "string",
"description": "Word paragraph style ID, e.g. Heading1 or Normal."
},
"paragraphs": {
"type": "array",
"description": "Batch structural DOCX paragraph insertions. Each item needs text and paragraph or anchor.",
"items": {
"type": "object",
"properties": {
"text": {
"type": "string"
},
"style": {
"type": "string"
},
"paragraph": {
"type": "integer",
"minimum": 1
},
"anchor": {
"type": "string"
},
"position": {
"type": "string",
"enum": [
"before",
"after"
]
}
},
"required": [
"text"
],
"additionalProperties": false
}
},
"paragraphTargets": {
"type": "array",
"description": "DOCX paragraph targets for deletion or style updates; use paragraph index or exact anchor text.",
"items": {
"type": "object",
"properties": {
"paragraph": {
"type": "integer",
"minimum": 1
},
"anchor": {
"type": "string"
}
},
"additionalProperties": false
}
},
"table": {
"type": "integer",
"minimum": 1,
"description": "1-based DOCX table number."
},
"row": {
"type": "integer",
"minimum": 1,
"description": "1-based DOCX table row."
},
"column": {
"type": "integer",
"minimum": 1,
"description": "1-based DOCX table column."
},
"tableCells": {
"type": "array",
"description": "Batch DOCX table updates. May accompany fill_template so placeholders and cells update in one version.",
"items": {
"type": "object",
"properties": {
"table": {
"type": "integer",
"minimum": 1
},
"row": {
"type": "integer",
"minimum": 1
},
"column": {
"type": "integer",
"minimum": 1
},
"text": {
"type": "string"
}
},
"required": [
"table",
"row",
"column",
"text"
],
"additionalProperties": false
}
},
"pages": {
"type": "array",
"items": {
"type": "integer",
"minimum": 1
},
"description": "Optional 1-based PDF pages; omitted means all pages."
},
"order": {
"type": "array",
"items": {
"type": "integer",
"minimum": 1
},
"description": "Complete 1-based page order for reorder_pages."
},
"degrees": {
"type": "integer",
"enum": [
90,
180,
270,
-90
]
},
"opacity": {
"type": "number",
"minimum": 0.05,
"maximum": 1
},
"title": {
"type": "string"
},
"subject": {
"type": "string"
},
"author": {
"type": "string",
"description": "PDF author metadata for set_properties."
},
"keywords": {
"type": "string",
"description": "Comma-separated keywords."
},
"values": {
"type": "object",
"additionalProperties": {
"type": "string"
},
"description": "For fill_template, maps placeholder names to replacement text; e.g. { client: 'Acme' } fills {{client}}."
},
"formValues": {
"type": "object",
"description": "For fill_form, maps PDF field names to text, boolean, or string-array values.",
"additionalProperties": {
"anyOf": [
{
"type": "string"
},
{
"type": "boolean"
},
{
"type": "array",
"items": {
"type": "string"
}
}
]
}
},
"flatten": {
"type": "boolean",
"description": "Flatten PDF form fields after filling."
},
"validate": {
"type": "boolean",
"description": "Validate the new version in the same call and return structural facts and warnings."
},
"sheet": {
"type": "string",
"description": "Optional XLSX worksheet name for set_cells."
},
"cells": {
"type": "array",
"description": "XLSX cell updates. Each item sets a value or formula in an A1 address.",
"items": {
"type": "object",
"properties": {
"address": {
"type": "string"
},
"value": {
"anyOf": [
{
"type": "string"
},
{
"type": "number"
},
{
"type": "boolean"
},
{
"type": "null"
}
]
},
"formula": {
"type": "string"
},
"result": {
"anyOf": [
{
"type": "string"
},
{
"type": "number"
},
{
"type": "boolean"
},
{
"type": "null"
}
]
},
"style": {
"type": "object",
"properties": {
"bold": {
"type": "boolean"
},
"italic": {
"type": "boolean"
},
"fontColor": {
"type": "string",
"description": "Hex RGB/ARGB color, e.g. #17365D."
},
"fillColor": {
"type": "string",
"description": "Solid fill hex RGB/ARGB color."
},
"numberFormat": {
"type": "string",
"description": "Excel number format, e.g. $#,##0.00 or 0.0%."
},
"horizontal": {
"type": "string",
"enum": [
"left",
"center",
"right"
]
},
"vertical": {
"type": "string",
"enum": [
"top",
"middle",
"bottom"
]
},
"wrapText": {
"type": "boolean"
}
},
"additionalProperties": false
}
},
"required": [
"address"
],
"additionalProperties": false
}
},
"sheetAction": {
"type": "string",
"enum": [
"create",
"rename",
"delete",
"configure"
],
"description": "XLSX worksheet lifecycle action."
},
"newSheetName": {
"type": "string",
"description": "New worksheet name for sheetAction=create or rename."
},
"freezeRows": {
"type": "integer",
"minimum": 0,
"description": "Number of top worksheet rows to freeze."
},
"freezeColumns": {
"type": "integer",
"minimum": 0,
"description": "Number of left worksheet columns to freeze."
},
"columnWidths": {
"type": "object",
"additionalProperties": {
"type": "number",
"minimum": 1
},
"description": "XLSX column widths keyed by A1 column letters or number."
},
"workbook": {
"type": "object",
"description": "Create a presentation-ready XLSX workbook. Use action=create_workbook; rows/cells are explicit local values.",
"properties": {
"name": {
"type": "string"
},
"sheets": {
"type": "array",
"minItems": 1,
"items": {
"type": "object",
"properties": {
"name": {
"type": "string"
},
"title": {
"type": "string",
"description": "Optional styled title row."
},
"headers": {
"type": "array",
"items": {
"type": "string"
},
"description": "Header labels written on headerRow."
},
"rows": {
"type": "array",
"items": {
"type": "array",
"items": {
"anyOf": [
{
"type": "string"
},
{
"type": "number"
},
{
"type": "boolean"
},
{
"type": "null"
}
]
}
}
},
"cells": {
"type": "array",
"items": {
"type": "object",
"properties": {
"address": {
"type": "string",
"description": "A1 cell address."
},
"value": {
"anyOf": [
{
"type": "string"
},
{
"type": "number"
},
{
"type": "boolean"
},
{
"type": "null"
}
]
},
"formula": {
"type": "string",
"description": "Excel formula without or with a leading equals sign."
},
"result": {
"anyOf": [
{
"type": "string"
},
{
"type": "number"
},
{
"type": "boolean"
},
{
"type": "null"
}
],
"description": "Optional cached/displayed formula result."
},
"style": {
"type": "object",
"properties": {
"bold": {
"type": "boolean"
},
"italic": {
"type": "boolean"
},
"fontColor": {
"type": "string",
"description": "Hex RGB/ARGB color, e.g. #17365D."
},
"fillColor": {
"type": "string",
"description": "Solid fill hex RGB/ARGB color."
},
"numberFormat": {
"type": "string",
"description": "Excel number format, e.g. $#,##0.00 or 0.0%."
},
"horizontal": {
"type": "string",
"enum": [
"left",
"center",
"right"
]
},
"vertical": {
"type": "string",
"enum": [
"top",
"middle",
"bottom"
]
},
"wrapText": {
"type": "boolean"
}
},
"additionalProperties": false
}
},
"required": [
"address"
],
"additionalProperties": false
}
},
"titleRow": {
"type": "integer",
"minimum": 1
},
"headerRow": {
"type": "integer",
"minimum": 1
},
"freezeRows": {
"type": "integer",
"minimum": 0
},
"freezeColumns": {
"type": "integer",
"minimum": 0
},
"columnWidths": {
"type": "object",
"additionalProperties": {
"type": "number",
"minimum": 1
}
},
"currencyColumns": {
"type": "array",
"items": {
"type": "string"
}
},
"percentageColumns": {
"type": "array",
"items": {
"type": "string"
}
},
"statusColumn": {
"type": "string"
},
"statusStyles": {
"type": "object",
"additionalProperties": {
"type": "object",
"properties": {
"bold": {
"type": "boolean"
},
"italic": {
"type": "boolean"
},
"fontColor": {
"type": "string",
"description": "Hex RGB/ARGB color, e.g. #17365D."
},
"fillColor": {
"type": "string",
"description": "Solid fill hex RGB/ARGB color."
},
"numberFormat": {
"type": "string",
"description": "Excel number format, e.g. $#,##0.00 or 0.0%."
},
"horizontal": {
"type": "string",
"enum": [
"left",
"center",
"right"
]
},
"vertical": {
"type": "string",
"enum": [
"top",
"middle",
"bottom"
]
},
"wrapText": {
"type": "boolean"
}
},
"additionalProperties": false
}
}
},
"required": [
"name"
],
"additionalProperties": false
}
}
},
"required": [
"name",
"sheets"
],
"additionalProperties": false
}
},
"required": [
"action"
],
"additionalProperties": false
}
}format_documentAction
Semantically rebuild the selected DOCX as a new version using docx + Mammoth. Choose a preset and optionally control page, margins, typography, spacing, alignment, headers, footers, and page numbers.
View tool JSON
{
"name": "format_document",
"kind": "act",
"impl": "imperative",
"description": "Semantically rebuild the selected DOCX as a new version using docx + Mammoth. Choose a preset and optionally control page, margins, typography, spacing, alignment, headers, footers, and page numbers.",
"inputSchema": {
"type": "object",
"properties": {
"fileId": {
"type": "string",
"description": "Optional file ID from workspace_file(list); omitted uses the selected file."
},
"preset": {
"type": "string",
"enum": [
"research",
"academic",
"business"
]
},
"pageSize": {
"type": "string",
"enum": [
"A4",
"letter"
]
},
"orientation": {
"type": "string",
"enum": [
"portrait",
"landscape"
]
},
"margins": {
"type": "object",
"description": "Margins in millimeters.",
"properties": {
"top": {
"type": "number",
"minimum": 0
},
"right": {
"type": "number",
"minimum": 0
},
"bottom": {
"type": "number",
"minimum": 0
},
"left": {
"type": "number",
"minimum": 0
}
},
"additionalProperties": false
},
"bodyFont": {
"type": "string"
},
"bodySize": {
"type": "number",
"minimum": 6,
"maximum": 72,
"description": "Points."
},
"lineSpacing": {
"type": "number",
"minimum": 0.8,
"maximum": 4
},
"paragraphAfter": {
"type": "number",
"minimum": 0,
"maximum": 72,
"description": "Points after paragraphs."
},
"alignment": {
"type": "string",
"enum": [
"left",
"center",
"right",
"justify"
]
},
"headerText": {
"type": "string"
},
"footerText": {
"type": "string"
},
"pageNumbers": {
"type": "boolean"
},
"validate": {
"type": "boolean",
"description": "Validate the formatted version in the same call."
}
},
"additionalProperties": false
}
}convert_fileAction
Create a derived local file from fileId. For datasets, applies an optional declarative DataFrame pipeline and exports CSV, JSON, JSONL, XLSX, or Parquet without Python. Also supports document/PDF/image conversions and inline validation.
View tool JSON
{
"name": "convert_file",
"kind": "act",
"impl": "imperative",
"description": "Create a derived local file from fileId. For datasets, applies an optional declarative DataFrame pipeline and exports CSV, JSON, JSONL, XLSX, or Parquet without Python. Also supports document/PDF/image conversions and inline validation.",
"inputSchema": {
"type": "object",
"properties": {
"fileId": {
"type": "string",
"description": "Optional file ID from workspace_file(list); omitted uses the selected file."
},
"format": {
"type": "string",
"enum": [
"pdf",
"docx",
"csv",
"json",
"jsonl",
"xlsx",
"parquet",
"webp",
"html",
"markdown",
"text"
]
},
"pages": {
"type": "array",
"items": {
"type": "integer",
"minimum": 1
},
"description": "For PDF→PDF, optional 1-based pages to extract."
},
"fileIds": {
"type": "array",
"items": {
"type": "string"
},
"description": "For PDF→PDF or DOCX→DOCX, file IDs to append in order."
},
"outputName": {
"type": "string",
"description": "Optional exact name for the derived output (save-as)."
},
"data": {
"type": "object",
"description": "Declarative DataFrame pipeline. Omit fields you do not need.",
"properties": {
"sheet": {
"type": "string",
"description": "Optional XLSX worksheet name for the selected dataset."
},
"joins": {
"type": "array",
"description": "Join other imported datasets by file ID. Use base.column and alias.column in select/filter/groupBy when needed.",
"items": {
"type": "object",
"properties": {
"fileId": {
"type": "string",
"description": "Dataset ID returned by workspace_file(list)."
},
"alias": {
"type": "string",
"description": "Optional source alias; defaults to joined_1, joined_2, and so on."
},
"sheet": {
"type": "string",
"description": "Optional XLSX worksheet name for this source."
},
"type": {
"type": "string",
"enum": [
"inner",
"left",
"right",
"full"
]
},
"leftColumn": {
"type": "string",
"description": "Column in the selected base dataset."
},
"rightColumn": {
"type": "string",
"description": "Column in the joined dataset."
}
},
"required": [
"fileId",
"leftColumn",
"rightColumn"
],
"additionalProperties": false
}
},
"unions": {
"type": "array",
"description": "Append rows from compatible imported datasets by column name.",
"items": {
"type": "object",
"properties": {
"fileId": {
"type": "string",
"description": "Dataset ID returned by workspace_file(list)."
},
"sheet": {
"type": "string",
"description": "Optional XLSX worksheet name for this source."
}
},
"required": [
"fileId"
],
"additionalProperties": false
}
},
"select": {
"type": "array",
"items": {
"type": "string"
},
"description": "Columns to project; qualified aliases are supported for joins."
},
"filters": {
"type": "array",
"items": {
"type": "object",
"properties": {
"column": {
"type": "string"
},
"operator": {
"type": "string",
"enum": [
"eq",
"neq",
"gt",
"gte",
"lt",
"lte",
"contains",
"starts_with",
"ends_with",
"in",
"is_null",
"not_null"
]
},
"value": {
"description": "Scalar value, or an array for the in operator."
}
},
"required": [
"column",
"operator"
],
"additionalProperties": false
}
},
"groupBy": {
"type": "array",
"items": {
"type": "string"
},
"description": "Grouping columns."
},
"aggregations": {
"type": "array",
"items": {
"type": "object",
"properties": {
"operation": {
"type": "string",
"enum": [
"count",
"sum",
"mean",
"min",
"max",
"median",
"nunique"
]
},
"column": {
"type": "string",
"description": "Omit only for count(*)."
},
"as": {
"type": "string",
"description": "Optional output column name."
}
},
"required": [
"operation"
],
"additionalProperties": false
}
},
"sort": {
"type": "array",
"items": {
"type": "object",
"properties": {
"column": {
"type": "string"
},
"direction": {
"type": "string",
"enum": [
"asc",
"desc"
]
}
},
"required": [
"column"
],
"additionalProperties": false
}
},
"offset": {
"type": "integer",
"minimum": 0
},
"limit": {
"type": "integer",
"minimum": 0
}
},
"additionalProperties": false
},
"validate": {
"type": "boolean",
"description": "Validate the derived artifact in the same call."
}
},
"required": [
"format"
],
"additionalProperties": false
}
}validate_fileAction
Validate a file's locally parsable structure by fileId, or use the selected file when omitted. Returns structural facts and explicit compatibility warnings.
View tool JSON
{
"name": "validate_file",
"kind": "act",
"impl": "imperative",
"description": "Validate a file's locally parsable structure by fileId, or use the selected file when omitted. Returns structural facts and explicit compatibility warnings.",
"inputSchema": {
"type": "object",
"properties": {
"fileId": {
"type": "string",
"description": "Optional file ID from workspace_file(list); omitted uses the selected file."
}
},
"additionalProperties": false
}
}render_previewAction
Bring the selected file preview into view and return its current preview model.
View tool JSON
{
"name": "render_preview",
"kind": "act",
"impl": "imperative",
"description": "Bring the selected file preview into view and return its current preview model.",
"inputSchema": {
"type": "object",
"properties": {},
"additionalProperties": false
}
}