office-document-mcp-server
A multi-capability MCP server for Office document automation, Azure cost lookups, and web research utilities.
Read, inspect, and audit Word, Excel, and PowerPoint documents.
Patch cells, placeholders, shapes, and sections with best-effort/safe/strict/dry-run modes.
Manage comments: add, get, reply, delete, resolve/reopen (Word threading).
Work with tables: get/add/update rows and create Word/PowerPoint tables.
Insert images into Word, Excel, and PowerPoint.
Copy/analyze templates and generate SOW documents from templates or Markdown.
Track changes in Word: enable tracking, apply tracked replacements, accept all changes.
Manipulate PowerPoint: add/delete/duplicate/reorder/hide slides, notes, import slides, recommend layouts.
Convert Markdown to Word, Excel, PowerPoint; create SOW from Markdown.
Excel extras: add sheets/charts, list sheets, delete comments.
Query Azure pricing: fetch/cache prices, calculate costs, list services/regions.
Web utilities: search, fetch pages, extract links/tables, check URLs.
Utility/tips: office_help workflow guidance, set comment identity, restart server, list supported formats.
Allows creating Word, Excel, and PowerPoint documents from Markdown content, and reading Office documents as Markdown.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@office-document-mcp-server Read the content of annual_report.docx"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Office Document MCP Server
This Python MCP server reads and edits unencrypted .docx, .xlsx, .xlsm and .pptx files, and can create .docx, .xlsx and .pptx files. It runs over local stdio by default; the pinned transport dependency also supports Streamable HTTP sessions and legacy HTTP/SSE or raw TCP. Legacy binary Office files, encrypted packages and Information Rights Management are unsupported.
Covered edits use a private copy, reopen the saved package and replace the destination once. Excel cell patches preserve style dependencies and invalidate cell-level formula caches; supported Word and PowerPoint literal replacements retain formatting across adjacent runs. Reopening does not verify rendered layout or recalculate formulas. See the operating limits and writer scope before editing complex files.
The project accepts patches and hotfixes but does not offer support or issue tracking.
Available Tools
Start with office_help when choosing a workflow. Use office_read or office_inspect before editing with office_patch, office_table, office_template, office_comment or office_image; use office_audit to check the result. word_insert_at_anchor handles paragraph placement when a literal target is unsuitable. The specialised tools below cover conversion, slides, tracked changes and older SOW templates.
Unified tools
These tools select document handlers from the file extension or provide cross-format guidance. office_set_comment_identity configures default comment attribution separately.
Tool | Description |
| Structured workflow help and recommendations for consulting/architecture document workflows |
| Read content from Word/Excel/PowerPoint as JSON or Markdown |
| Get document structure (sheets, slides, sections, tables, comments) |
| Edit cells, shapes, sections, or replace placeholders |
| Add/get/reply/delete comments; Word also supports resolve/reopen, threaded get, and reply threading |
| Read tables, add/update rows, and create Word/PowerPoint tables |
| Copy templates or analyze template structure |
| Audit for placeholders, completion, or tracking status |
| Insert images into Word, Excel, or PowerPoint documents |
Specialised tools
Use these for operations that the unified tools do not expose, or when an existing integration calls them directly.
Word SOW Generation
The sow tools began as a template-specific proof of concept. They are deprecated but remain available for existing integrations; use the unified tools for new editing workflows.
Tool | Description |
| Fill SOW template with structured data |
| Remove template artifacts and guidance (tracked) |
| Extract template instructions from a section |
| Analyze SOW template structure |
| Create SOW from Markdown content |
| Extract structured data from existing SOW |
| Insert paragraphs before/after an anchor paragraph or paragraph index |
| List headings and high-signal paragraphs that can be used as insertion anchors |
| Return a lightweight map of sections, tables, placeholders, anchors, and warnings |
| Enable Word's track changes mode |
| Replace text with revision marks |
| Accept insertion/deletion wrappers in the main document XML; see revision limits below |
PowerPoint Slide Management
Tool | Description |
| Add new slide with specified layout |
| Remove a slide |
| Copy a slide with independent chart and embedded-workbook parts |
| Change slide order |
| Hide/unhide a slide |
| Set speaker notes |
| Get best layout for content type |
| Add change log slide |
| Copy a slide between presentations, with optional speaker notes |
Document Conversion
Tool | Description |
| Create Word document from Markdown (supports inline text or |
| Create Excel workbook from Markdown tables (supports inline text or |
| Create PowerPoint from Markdown slides (supports inline text or |
Utility
Tool | Description |
| Hot-reload the server after code changes |
| Show available document formats |
Related MCP server: DOCX-MCP
Quick Examples
Protocol guidance and results
Tool results keep their legacy text content and add structuredContent for mappings. Failed operations set MCP isError; partial success still needs per-target checks. Tool lists support pagination and explicit conservative annotations. The read-only office://guidance/workflows resource and review_document prompt offer workflow guidance without opening files; the prompt's document-type argument supports completion.
See the uMCP integration for opt-in progress, cooperative cancellation, HTTP authentication/session rules and the pinned dependency update procedure.
Workflow Discovery
# Find the best workflow for filling a consulting SOW from markdown
office_help(
goal="fill_sow_from_markdown",
document_type="word",
constraints=["preserve_template_structure"],
format="summary"
)
# Map a common consulting request onto a deterministic workflow
office_help(
task="Patch an Excel estimate workbook safely and verify the result",
format="detailed"
)
# Discover the safest path for a stakeholder review deck
office_help(
goal="create_review_deck",
document_type="powerpoint",
format="summary"
)Template Analysis Cache
word_parse_sow_template and office_template(operation="analyze") cache Word template metadata as JSON in .office-metadata-cache/. Set OFFICE_MCP_METADATA_CACHE_DIR to move the cache. A resolved-path, file-size or mtime change invalidates an entry.
Reading Documents
# Read Excel as Markdown
office_read(file_path="data.xlsx", output_format="markdown")
# Read specific range
office_read(file_path="data.xlsx", scope="Sheet1!A1:D10")
# Read a single worksheet
office_read(file_path="data.xlsx", scope="Sheet1")
# Read Excel formulas instead of cached values (reading never recalculates)
office_read(file_path="model.xlsx", include_formulas=True)
# Read Word document
office_read(file_path="report.docx", output_format="markdown")Inspecting Structure
# List Excel sheets
office_inspect(file_path="data.xlsx", what="sheets")
# List Word tables
office_inspect(file_path="report.docx", what="tables")
# List PowerPoint slides
office_inspect(file_path="deck.pptx", what="slides")
# Analyze a Word template and reuse cached metadata on later runs
office_template(
source_path="templates/sow.docx",
destination_path="",
operation="analyze"
)
# Discover insertion anchors before adding narrative content
word_list_anchors(file_path="report.docx", query="delivery")
# Get a compact map of a Word document
word_document_map(file_path="report.docx")Editing Content
# Patch Excel cell
office_patch(
file_path="data.xlsx",
changes=[{"target": "A1", "value": "New Value"}]
)
# Patch Word placeholder
office_patch(
file_path="report.docx",
changes=[{"target": "<Customer>", "value": "Contoso"}]
)
# Patch PowerPoint shape
office_patch(
file_path="deck.pptx",
changes=[{"target": "slide:1/Title 1", "value": "New Title"}]
)
# PowerPoint soft return in a single text box
office_patch(
file_path="deck.pptx",
changes=[{"target": "slide:1/Title 2", "value": "Contoso{br}Project"}]
)Covered mutation tools return per-target diagnostics. office_patch, office_table and office_comment accept best_effort, safe, strict and dry_run; other tools accept a mode only if it appears in their schema. Fields depend on the tool and how early validation fails:
successstatus(success,partial_success,failed,skipped)warningsmatched_targetsunmatched_targetsskipped_targetsdiagnosticsnext_toolsbest_effort: current compatibility-oriented behaviorsafe: requires a distinct output path for covered mutation flowsstrict: foroffice_patch, any missing or failed target prevents the entire batch commitdry_run: foroffice_patch, validates changes on a private copy, discards it, and leaves source and destination unchanged
For office_patch, safe requires a distinct destination but may commit the accepted subset. Use strict with a new output_path when every target is required. Receipts distinguish changes_planned from committed changes_applied, include results[].applied, and record source_sha256.
Source/destination fingerprint changes prevent publication. Process-local writer locks cover staged patches and the enrolled existing-document writers in writer scope; arbitrary external editors are not locked. Hard-linked document paths refuse mutation. Tables and comments share staged publication, but their XLSX serialisation does not preserve every opaque part as office_patch does. Output-only generation paths retain separate contracts.
Excel cell patches preserve append-only style dependencies and invalidate formula caches on cell-level formula elements across worksheets. calculation_state="recalculation-required" means an external calculation engine must refresh results; no recalculation runs here. Unsupported style registry rewrites refuse before commit. Word read-back includes tracked insertions and excludes tracked deletions.
Start with office_help, then inspect, preview, patch to a new output, and reopen/audit the result. Excel value patches coerce numeric-looking strings, currencies and percentages; this API has no force-text option for numeric identifiers. See operating limits before processing untrusted or complex files.
Verification
Install the development dependencies before running bash tests/run_tests.sh, or select related test paths as arguments. Set PYTHON=/path/to/python to choose the environment. The testing guide covers setup, focused batches, Gherkin reports and the optional LibreOffice checks. Verification never auto-formats or fixes source.
Shared acceptance scenarios execute through pytest-bdd in tests/acceptance/. Each run replaces test-results/acceptance.json with a fresh inventory and per-step outcomes. Planned, undefined, ambiguous and unexecuted cases cannot count as acceptance passes. Fixtures and shared requirements come from the tagged references/fixtures-ooxml submodule. Python reads the shared Gherkin directly and uses tests/acceptance/shared-mapping.json for local implementation status; no duplicate executable feature copy is maintained. Schema-2 document inputs have stable content IDs and one physical payload under central fixtures/<format>/<scenario-group>/; tests obtain paths from the manifest. The separate native-test catalogue staging directory is incomplete reconciliation input, not additional executable coverage.
The testing guide separates historical runtime results, fixture-migration measurements and current release checks. The uMCP integration report records results at its original source pin, including an official-SDK smoke check. See the testing guide for the current fixture release and its bounded checks. Optional local LibreOffice tests skipped when the executable was unavailable; the manual oracle workflow installs it for separate checks. Native Microsoft Office rendering and Windows executable mutation workflows have not been verified locally.
Recorded test results distinguish committed tests from local-only tests. The XLSX adoption decision explains why the server retains upstream openpyxl.
Table Operations
# Add row to Word table
office_table(
file_path="report.docx",
operation="add_row",
table_id="staffing",
data={"Role": "PM", "Count": "1", "Notes": "Lead"}
)
# Create table in Word
office_table(
file_path="report.docx",
operation="create",
data={
"headers": ["Phase", "Owner", "Target Date"],
"rows": [{"Phase": "Discovery", "Owner": "PM", "Target Date": "2026-04-01"}],
"insert_after_section": "Delivery Plan"
}
)
# Add table to PowerPoint
office_table(
file_path="deck.pptx",
operation="create",
table_id="3",
data={
"headers": ["Phase", "Duration"],
"rows": [["Discovery", "2 weeks"]]
}
)Track Changes Workflows
# Enable Track Changes in document settings
word_enable_track_changes(file_path="draft.docx", output_path="draft-tracked.docx")
# Apply a tracked replacement
word_patch_with_track_changes(
file_path="draft-tracked.docx",
replacements={"Old wording": "New wording"},
output_path="draft-review.docx"
)
# Accept main-document insertion/deletion wrappers
word_accept_all_changes(
file_path="draft-review.docx",
output_path="draft-final.docx"
)word_accept_all_changes traverses the main document XML, including its tables. It does not resolve revisions in separate headers, footers, footnotes or other story parts, and it does not implement move/format revision resolution or reject-all. Despite its name, it is not a complete Word revision engine. The office_patch(track_changes=...) compatibility parameter currently does not control Word dispatch; use the documented tracked workflow and inspect its output.
Image Support
office_image supports raster formats (PNG, JPG/JPEG, GIF) across Word, Excel, and PowerPoint.
SVG support is currently:
Format | Word | Excel | PowerPoint |
PNG | Yes | Yes | Yes |
SVG | Yes | No | Yes |
Notes:
Excel image placement is cell-anchored.
PowerPoint image placement is centered by default and can be validated from returned position metadata.
Recent local MCP validation covered image insertion plus position/bounds checks for Excel and PowerPoint fixture copies.
Large Markdown Inputs
# Avoid MCP argument-size limits by passing a markdown_file path
word_from_markdown(
output_path="report.docx",
markdown_file="inputs/large-report.md"
)
excel_from_markdown(
output_path="budget.xlsx",
markdown_file="inputs/budget-tables.md"
)
pptx_from_markdown(
output_path="deck.pptx",
markdown_file="inputs/deck.md"
)
word_create_sow_from_markdown(
output_path="sow.docx",
template_path="templates/Agile.docx",
markdown_file="inputs/sow.md"
)Auditing
# Audit for placeholders
office_audit(file_path="report.docx", checks=["placeholders"])
# Audit for completion
office_audit(file_path="report.docx", checks=["completion"])Comment Threads and Resolution (Word)
Word stores comment text in word/comments.xml and thread/resolution metadata in word/commentsExtended.xml.
word_get_comments
word_get_comments(
file_path="SoW.docx",
filter="all", # all | open | resolved | mine
author=None, # used with filter="mine"
format="flat" # flat | threaded
)Each comment now includes:
idauthorinitialsdatetextdone(resolved state)is_replyparent_idpara_id
When format="threaded", response includes:
threads: grouped{ root, replies[] }flat: full backward-compatible flat list
word_resolve_comment
word_resolve_comment(
file_path="SoW.docx",
comment_id="121",
resolved=True, # True=resolve, False=reopen
output_path=None,
)Notes:
If a reply ID is supplied, the root thread is resolved/reopened.
If
commentsExtended.xmlis missing, it is created and wired into the package.If root comments lack
w14:paraId, a paraId is synthesized for stable mapping.
word_reply_to_comment
word_reply_to_comment(
file_path="SoW.docx",
comment_id="121",
text="Done — updated as requested.",
author="Rui Carmo",
auto_resolve=True,
)auto_resolve=True performs reply + resolve in one call.
Unified API (office_comment)
office_comment supports Word thread workflows directly:
office_comment(file_path="SoW.docx", operation="get", format="threaded")
office_comment(file_path="SoW.docx", operation="get", filter="open")
office_comment(file_path="SoW.docx", operation="resolve", target="121")
office_comment(file_path="SoW.docx", operation="reopen", target="121")Supported operations:
Word:
add,get,reply,resolve,reopen,deleteExcel:
add,get,delete(reply/resolve/reopenreturn clear unsupported errors)PowerPoint:
add,get,delete(reply/resolve/reopenreturn clear unsupported errors)
End-to-end round-trip tests
Fixture-based tests cover:
mixed open/resolved states
resolve/reopen toggling
reply-to-root resolution behaviour
legacy
commentsIds.xmlfallbackmissing
commentsExtended.xmlcreationthreaded retrieval + filtering
output-path round trips
Primary test files:
tests/test_word_comment_resolution.pytests/test_word_comment_roundtrip_fixture.pytests/test_word_comment_replies.py
Word Review Workflow
Inspect the document before choosing targets. Use section: targets for section content and literal text for placeholders; office_patch has no operation argument.
office_inspect(file_path="draft.docx", what="sections")
word_list_anchors(file_path="draft.docx", query="Delivery")
changes = [
{"target": "<Customer>", "value": "Contoso"},
{"target": "section:Delivery", "value": "Delivery starts after approval."}
]
office_patch(file_path="draft.docx", changes=changes, mode="dry_run")
office_patch(
file_path="draft.docx", changes=changes,
mode="strict", output_path="review.docx"
)
office_read(file_path="review.docx")
office_audit(file_path="review.docx", checks=["placeholders", "completion"])Preview reports planned changes; strict commit requires every requested target to apply. word_patch_with_track_changes writes tracked replacements; office_patch does not route its track_changes argument into Word editing. Read-back includes tracked insertions and excludes deletions. Inspect the saved document before accepting revisions; package checks cannot establish its meaning or rendered layout.
Setup
Install from repository (recommended)
Using uv:
uv tool install "git+https://github.com/rcarmo/python-office-mcp-server.git"Using pip:
python -m venv .venv
. .venv/bin/activate # Windows PowerShell: .venv\\Scripts\\Activate.ps1
python -m pip install "git+https://github.com/rcarmo/python-office-mcp-server.git"This installs the office-mcp-server command and the exact Git-pinned transport dependency. Git must be available during installation. The public PyPI package named umcp is unrelated; use the repository's dependency declaration. Requires Python >=3.10; locally tested on 3.10, 3.12 and 3.13. If a GUI client cannot find the command, configure the absolute executable path; its environment may not inherit your shell's PATH.
Install from local clone
git clone --recurse-submodules https://github.com/rcarmo/python-office-mcp-server.git
cd python-office-mcp-server
uv sync --frozen
uv run office-mcp-server
# For development and tests: uv sync --frozen --extra dev
# In an activated pip environment: python -m pip install -e '.[dev]'Run the source script after installing dependencies
git clone --recurse-submodules https://github.com/rcarmo/python-office-mcp-server.git
cd python-office-mcp-server
python -m venv .venv
. .venv/bin/activate # Windows PowerShell: .venv\\Scripts\\Activate.ps1
python -m pip install -r requirements.txt
python office_server.pyThis starts the stdio server; it waits for MCP messages rather than opening a web page. Configure the client with the same virtual-environment interpreter and an absolute script path:
{
"command": "/absolute/path/to/python-office-mcp-server/.venv/bin/python",
"args": ["/absolute/path/to/python-office-mcp-server/office_server.py"]
}On Windows, use .venv\\Scripts\\python.exe and escape backslashes in JSON. Relative document paths depend on the server process's working directory; absolute paths avoid ambiguity.
For network clients use explicit --http; plain --port retains legacy SSE. Configure OFFICE_MCP_HTTP_TOKEN through the server environment for bearer authentication, and keep a TLS proxy and OS/path boundary for remote access. See transports. The client examples below use stdio.
MCP client configuration
VS Code (.vscode/mcp.json):
{
"servers": {
"office": {
"command": "office-mcp-server"
}
}
}Claude Desktop (claude_desktop_config.json):
{
"mcpServers": {
"office": {
"command": "office-mcp-server"
}
}
}mcp-cli (mcp_servers.json):
{
"mcpServers": {
"office": {
"command": "office-mcp-server"
}
}
}To use uvx instead of a pre-installed binary:
{
"command": "uvx",
"args": ["--from", "git+https://github.com/rcarmo/python-office-mcp-server.git", "office-mcp-server"]
}Tool Activation Note
Some MCP clients can start with subsets of tools disabled by policy/session settings. If a call returns a disabled-tool error, enable the corresponding MCP tools in the client first, then retry. This enablement behavior is controlled by the MCP client/host, not by this server.
VS Code (Automatic)
Create .vscode/mcp.json using the VS Code example above; the repository does not ship that file. Open the Command Palette and run MCP: List Servers to confirm office is listed.
GitHub Copilot CLI
Add the server to your Copilot CLI configuration:
# Open config file
code ~/.config/github-copilot/config.json
# Add this to the mcpServers section:
{
"mcpServers": {
"officeServer": {
"command": "python",
"args": ["/path/to/python-office-mcp-server/office_server.py"]
}
}
}Running Manually
cd python-office-mcp-server
pip install -r requirements.txt
python office_server.pyWindows Single-File Distribution
Build a standalone .exe using PyInstaller.
Build on Windows
cd python-office-mcp-server
python -m pip install -r requirements.txt
python -m pip install -r requirements-build.txt
python build_windows_onefile.py --cleanOutput artifact:
dist/office-mcp-server.exe
Custom output name
python build_windows_onefile.py --name office-server-prodRun executable
dist\office-mcp-server.exeUse the generated executable in MCP client configuration by pointing command to the .exe path.
The Windows workflow builds the executable and checks tool discovery. It does not run mutation workflows through the executable.
Dependencies
python-docx— Word document handlingopenpyxl— Excel workbook handlingpython-pptx— PowerPoint presentation handlingreferences/fixtures-ooxml— Git submodule of tagged fixtures and behaviour contracts used by testsInstalled uMCP 0.2.2 dependency — Async transport and request context; exact Git revision declared in project metadata
pyinstaller— Build-time dependency for one-file Windows executable
Architecture
The server dynamically loads tool modules from tools/:
office_unified_tools.py— Unified document operationsword_tools.py— Word conversion toolsword_advanced_tools.py— Revisions, anchors, tables and legacy SOW workflowsexcel_tools.py— Excel conversion toolsexcel_advanced_tools.py— Excel advanced operations (internal)pptx_tools.py— PowerPoint conversion toolspptx_advanced_tools.py— Slide management toolspptx_slide_transfer_tools.py— Relationship-aware slide importmcp_features.py— Office metadata, structured failures, guidance and HTTP/request policymutation.py— Staging, writer locks, fingerprints, cooperative cancellation and commit receiptspackage_guard.py/package_preservation.py— Bounded admission and package differencesxlsx_preservation.py— Cell-edit style and calculation dependenciesword_spans.py/pptx_text.py— Adjacent-run text replacement
Tools are discovered automatically by class name pattern (*Tools).
Available Tools
61 toolsazure_calculate_costARead-only
Calculate monthly cost for an Azure resource.
Looks up pricing and calculates the estimated monthly cost based on quantity and usage hours.
Example: # Cost for 3 D4 v5 VMs running 24/7 azure_calculate_cost( service="Virtual Machines", sku_match="D4 v5", quantity=3 )
# Cost for 1000 GB storage
azure_calculate_cost(
service="Storage",
product_match="Blob Storage",
sku_match="Hot LRS",
quantity=1000,
hours_per_month=1 # Storage is per GB, not per hour
)Args: service: Azure service name region: ARM region name (default: "westeurope") sku_match: SKU name to match product_match: Product name to match quantity: Number of units (VMs, instances, GB, etc.) hours_per_month: Hours of usage per month (default: 730 = 24/7) price_type: Price type (default: "Consumption") currency: Currency code (default: "USD")
Returns: Dictionary with pricing details and calculated costs
Inputs: service (required), region, sku_match, product_match, +4 more.
azure_clear_cacheADestructive
Clear Azure pricing cache.
Args: service: Clear cache for specific service only (None = all) region: Clear cache for specific region only (None = all) clear_disk: Clear disk cache (default: True) clear_memory: Clear memory cache (default: True)
Returns: Dictionary with cleared cache information
Inputs: service, region, clear_disk, clear_memory.
azure_fetch_pricesARead-only
Fetch and cache Azure retail prices for specified services and regions.
Downloads pricing data from the Azure Retail Prices API and stores it locally for fast subsequent queries. Data is cached on disk for 24 hours.
Example: azure_fetch_prices( services=["Virtual Machines", "Azure Databricks", "Storage"], regions=["westeurope", "eastus"] )
azure_fetch_prices(services=["API Management"], force_refresh=True)Args: services: List of Azure service names to fetch pricing for. If None, fetches common services. regions: List of ARM region names (e.g., "westeurope", "eastus"). If None, fetches common regions. currency: Currency code (default: "USD") force_refresh: If True, ignore cache and fetch fresh data
Returns: Dictionary with fetch status and summary statistics
Inputs: services, regions, currency, force_refresh.
azure_list_cached_servicesARead-only
List all services currently cached in memory and on disk.
Returns: Dictionary with cached service/region combinations and statistics
Inputs: no named parameters.
azure_list_regionsARead-only
List available ARM regions for a service or cached data.
Args: service: Optional service name to filter by currency: Currency code (default: "USD") from_cache_only: If True, only uses cached data max_pages: Maximum API pages to scan when fetching (default: 2) max_regions: Maximum regions to return (default: 200)
Returns: Dictionary with region names and source metadata
Inputs: service, currency, from_cache_only, max_pages, +1 more.
azure_list_servicesARead-only
List available Azure service names for a region.
Args: region: ARM region name (default: "westeurope") currency: Currency code (default: "USD") from_cache_only: If True, only uses cached data max_pages: Maximum API pages to scan when fetching (default: 2) max_services: Maximum services to return (default: 200)
Returns: Dictionary with service names and source metadata
Inputs: region, currency, from_cache_only, max_pages, +1 more.
azure_query_pricesARead-only
Query cached Azure pricing data with flexible filters.
Searches the in-memory price cache for matching items. If data is not cached, it will be fetched automatically.
Example: # Get VM pricing azure_query_prices( service="Virtual Machines", region="westeurope", sku_contains="D4" )
# Get Databricks DBU pricing
azure_query_prices(
service="Azure Databricks",
sku_contains="All-purpose"
)
# Get reserved instance pricing
azure_query_prices(
service="Virtual Machines",
sku_contains="D4",
price_type="Reservation"
)Args: service: Azure service name (required for initial query) region: ARM region name (default: "westeurope") sku_contains: Filter by SKU name containing this string product_contains: Filter by product name containing this string price_type: Price type filter: "Consumption", "Reservation", "DevTestConsumption", or None for all currency: Currency code (default: "USD") max_results: Maximum items to return (default: 50)
Returns: Dictionary with matching price items and summary
Inputs: service (required), region, sku_contains, product_contains, +4 more.
excel_add_chartA
Add a chart to an Excel worksheet.
Example: excel_add_chart( file_path="report.xlsx", data_range="Sheet1!A1:D8", chart_type="line", title="Growth Forecast", position="F2" )
Args: file_path: Path to the .xlsx or .xlsm file data_range: Range containing data (e.g., "A1:D10" or "Sheet1!A1:D10") chart_type: "line", "bar", "column", or "pie" sheet_name: Optional sheet name (overrides sheet in data_range) title: Optional chart title position: Top-left anchor cell for the chart (default "E2") has_header: Treat first row as header for series names use_first_column_as_categories: Use first column as category labels output_path: Optional output path (defaults to overwriting input)
Returns: Status dictionary with chart details
Inputs: file_path (required), data_range (required), chart_type, sheet_name, +5 more.
excel_add_sheetA
Add a new sheet to an Excel workbook.
Creates a new empty sheet in the workbook at the specified position. The sheet can be inserted at the start, end, or after a specific existing sheet.
Example: excel_add_sheet(file_path="data.xlsx", sheet_name="Summary") excel_add_sheet(file_path="data.xlsx", sheet_name="NewSheet", position="start") excel_add_sheet(file_path="data.xlsx", sheet_name="Details", position="Sheet1")
Args: file_path: Path to the .xlsx or .xlsm file sheet_name: Name for the new sheet position: Where to insert - 'start', 'end' (default), or name of sheet to insert after output_path: Optional output path (defaults to overwriting input file)
Returns: Dictionary with success status and sheet details
Inputs: file_path (required), sheet_name (required), position, output_path.
excel_delete_commentADestructive
Delete a comment from a specific cell.
Args: file_path: Path to the .xlsx or .xlsm file cell_ref: Cell reference (e.g., 'B5', 'Sheet1!C10') sheet_name: Optional sheet name (overrides sheet in cell_ref) output_path: Optional output path (defaults to overwriting input)
Returns: Status dictionary with deletion details
Inputs: file_path (required), cell_ref (required), sheet_name, output_path.
excel_from_markdownA
Convert Markdown tables to an Excel workbook from inline content or markdown_file.
This is the primary tool for creating Excel workbooks from text content.
Parses GitHub Flavored Markdown content and extracts all tables.
Each table becomes a separate sheet in the workbook.
Features:
- Auto-detects multiple tables in the content using GFM parser
- Header row gets bold formatting with gray background
- Auto-filter on header row
- Column widths auto-sized based on content
- Numbers and percentages are coerced to numeric types
- Supports inline formatting in cells (bold, italic, code)
- Supports formulas when a cell starts with '='
- Uses nearby '##' headings to name sheets (up to 31 chars)
Example:
excel_from_markdown(
output_path="04. Artifacts/budget.xlsx",
markdown='''
Category | Q1 | Q2 | Q3 | Q4 | Total |
Personnel | $50,000 | $52,000 | $54,000 | $56,000 | $212,000 |
Infrastructure | $15,000 | $15,000 | $16,000 | $16,000 | $62,000 |
Software | $8,000 | $8,500 | $9,000 | $9,500 | $35,000 |
Training | $5,000 | $3,000 | $4,000 | $3,000 | $15,000 |
Milestone | Target Date | Owner | Status |
Phase 1 Complete | 2026-03-31 | Alice | 100% |
Phase 2 Complete | 2026-06-30 | Bob | 45% |
Go-Live | 2026-09-30 | Carol | 0% |
''' |
)
Args:
output_path: Path for the output .xlsx file
markdown: GitHub Flavored Markdown content containing one or more tables (inline)
sheet_name: Optional sheet name for the first/only sheet
markdown_file: Optional path to a Markdown file. Use this for
very large inputs to avoid MCP argument-size limits.
Returns:
Status dictionary with file path and sheet count
Inputs: output_path (required), markdown, sheet_name, markdown_file.
excel_list_sheetsARead-only
List all sheets in an Excel workbook with their properties.
Provides detailed information about each sheet including visibility, dimensions, and whether it contains tables or data validations.
Example: excel_list_sheets(file_path="template.xlsx")
Args: file_path: Path to the .xlsx or .xlsm file include_hidden: Include hidden and very hidden sheets (default True)
Returns: Dictionary with sheet information including names, states, and metadata
Inputs: file_path (required), include_hidden.
list_supported_formatsARead-only
List supported document formats and their availability.
Returns: Dictionary showing which formats are available
Inputs: no named parameters.
office_auditARead-only
Audit documents for completeness, placeholders, and issues.
Replaces: excel_audit_placeholders, word_audit_completion, word_audit_sow, pptx_audit_placeholders
Examples: # Check for unfilled placeholders office_audit(file_path="contract.docx", checks=["placeholders"])
# Full completion audit
office_audit(file_path="sow.docx", checks=["completion"])
# Check Excel for placeholders
office_audit(file_path="estimate.xlsx", checks=["placeholders"])
# Multiple checks
office_audit(
file_path="document.docx",
checks=["placeholders", "tracking"]
)Args: file_path: Path to the document checks: List of checks to perform: - "placeholders": Find unfilled <...>, [...], [TBD] patterns - "completion": Full completion audit (Word SOW) - "tracking": Check for pending track changes - "formatting": Check for formatting issues - "empty_cells": Check required Excel cells for empty values - "totals": Verify Excel totals based on configured ranges - "dates": Validate Excel date formats (MM/DD/YYYY) audit_config: Optional configuration for Excel checks: - required_cells: list of cell refs to check for empty values - date_cells: list of cell refs to validate as MM/DD/YYYY - totals: list of dicts with sum_range, target, and optional tolerance
Returns: Dictionary with audit findings
Inputs: file_path (required), checks, audit_config.
office_commentA
Manage comments in Word, Excel, or PowerPoint documents.
Replaces: excel_add_comment, excel_get_comments, word_add_comment, pptx_add_comment, pptx_get_comments
Examples: # Get all comments from Excel office_comment(file_path="data.xlsx", operation="get")
# Add comment to Excel cell
office_comment(
file_path="data.xlsx",
operation="add",
target="B5",
text="Review this value"
)
# Add comment to Word text
office_comment(
file_path="report.docx",
operation="add",
target="project timeline",
text="Verify dates with PM"
)
# Reply to an existing Word comment by comment ID
office_comment(
file_path="report.docx",
operation="reply",
target="12", # comment ID from office_comment(..., operation="get")
text="Acknowledged - updated in v2"
)
# Add comment to PowerPoint slide
office_comment(
file_path="deck.pptx",
operation="add",
target="3", # or "slide:3"
text="Update chart data"
)Args: file_path: Path to the document operation: add/get/reply/delete plus resolve/reopen for Word target: Target location/ID - Excel add/delete: cell reference (e.g., "B5") - Excel get: optional sheet name filter - Word add: text span to annotate - Word reply/resolve/reopen/delete: comment ID - PowerPoint add: slide number ("3" or "slide:3") - PowerPoint delete: "slide:N" or "slide:N/comment:I" text: Comment text (required for add/reply) author: Author display name. Also used by Word get(filter="mine") output_path: Optional output path (defaults to overwriting input) format: For Word get only: "flat" (default) or "threaded" filter: For Word get only: "all" (default), "open", "resolved", "mine"
Returns: Dictionary with operation results
Inputs: file_path (required), text (required), operation, target, +4 more.
office_helpA
Get structured workflow help and recommendations for office document work.
Use this as the preferred discovery entry point for systems architecture
and consulting workflows. Prefer goal plus optional document_type
and constraints. task is supported only as a thin convenience layer
for mapping common natural-language requests onto the structured workflow
catalog.
Inputs: goal, document_type, constraints, task, +1 more.
office_imageA
Insert an image into a Word, Excel, or PowerPoint document.
Auto-detects document format from file extension and inserts the image at the specified location. Supports PNG, JPG/JPEG, and GIF formats.
Examples: # Insert image at end of Word document office_image( file_path="report.docx", image_path="logo.png", width_inches=2.0 )
# Insert image at specific paragraph in Word
office_image(
file_path="report.docx",
image_path="chart.png",
target="after:Executive Summary",
width_inches=5.0
)
# Insert image in Excel cell
office_image(
file_path="data.xlsx",
image_path="logo.png",
target="A1",
width_inches=1.5
)
# Insert image on specific Excel sheet
office_image(
file_path="data.xlsx",
image_path="chart.png",
target="Sheet2!B5"
)
# Insert image on PowerPoint slide
office_image(
file_path="deck.pptx",
image_path="diagram.png",
target="slide:2",
width_inches=4.0,
height_inches=3.0
)Args: file_path: Path to the document (.docx, .xlsx, .pptx) image_path: Path to the image file (.png, .jpg, .jpeg, .gif) target: Where to insert the image: - Word: "after:Section Title" or "end" (default) - Excel: cell reference like "A1" or "Sheet1!B5" - PowerPoint: "slide:N" where N is slide number (1-based) width_inches: Image width in inches (height auto-scales if not set) height_inches: Image height in inches (width auto-scales if not set) output_path: Optional output path (defaults to overwriting input)
Returns: Dictionary with insertion result
Inputs: file_path (required), image_path (required), target, width_inches, +2 more.
office_inspectARead-only
Inspect document structure and metadata.
Auto-detects document format and returns requested structural information.
Replaces: excel_list_sheets, excel_list_tables, excel_list_named_ranges, excel_list_merged_cells, excel_get_comments, excel_get_change_log, word_list_sections, word_list_tables, word_check_tracking, pptx_list_slides, pptx_list_shapes, pptx_list_masters, pptx_get_notes, pptx_get_comments, pptx_get_hidden_slides
Examples: # List all sheets in Excel workbook office_inspect(file_path="data.xlsx", what="sheets")
# List tables in Excel
office_inspect(file_path="data.xlsx", what="tables")
# Get comments from Excel
office_inspect(file_path="data.xlsx", what="comments")
# List slides in PowerPoint
office_inspect(file_path="deck.pptx", what="slides")
# List sections in Word
office_inspect(file_path="report.docx", what="sections")
# Get shapes on a specific slide
office_inspect(file_path="deck.pptx", what="shapes", target="3")Args: file_path: Path to the document what: What to inspect: - "structure": Overview of document structure - "sheets": Excel sheets list - "slides": PowerPoint slides list - "sections": Word sections list - "tables": Tables in document - "named_ranges": Excel named ranges - "merged_cells": Excel merged cell regions - "comments": Comments/notes in document - "tracking": Track changes status (Word) - "shapes": Shapes on a slide (PowerPoint) - "masters": Slide masters (PowerPoint) target: Optional target for scoped inspection: - Sheet name for Excel - Slide number for PowerPoint
Returns: Dictionary with inspection results
Inputs: file_path (required), what, target.
office_patchA
Apply edits to Word, Excel, or PowerPoint documents.
Accepts a list of changes and applies them to the document. Each change specifies a target (cell, shape, placeholder, section) and new value.
Replaces: excel_patch_cell, excel_patch_range, excel_replace_placeholders, word_patch_section, word_patch_placeholder, word_fix_split_placeholders, word_replace_global_variables, pptx_patch_shape, pptx_replace_text, pptx_replace_placeholders
Examples: # Patch Excel cells office_patch( file_path="data.xlsx", changes=[ {"target": "B5", "value": "New Value"}, {"target": "C10", "value": 42}, {"target": "D1", "value": "=SUM(A1:A10)"}, ] )
# Patch Excel range (multiple cells at once)
office_patch(
file_path="data.xlsx",
changes=[{"target": "A1:B3", "value": [["H1", "H2"], ["A", 1], ["B", 2]]}]
)
# Patch cells on a specific sheet (quote sheet names with special chars)
office_patch(
file_path="form.xlsm",
changes=[
{"target": "'ECIF Work Scope (E)'!B5", "value": "Contoso Ltd"},
{"target": "'ECIF Work Scope (E)'!B28", "value": "02/01/2026"},
]
)
# Replace placeholders in Word
office_patch(
file_path="template.docx",
changes=[
{"target": "<Customer Name>", "value": "Acme Corp"},
{"target": "<Date>", "value": "2026-01-23"},
]
)
# Patch PowerPoint shape
office_patch(
file_path="deck.pptx",
changes=[{"target": "slide:1/Title 1", "value": "New Title"}]
)
# Patch PowerPoint with soft return
office_patch(
file_path="deck.pptx",
changes=[{"target": "slide:1/Title 2", "value": "Contoso{br}Project"}]
)IMPORTANT for PowerPoint: When patching content placeholders (body, Content Placeholder), do NOT include bullet characters (•, -, *, etc.) in text lines. PowerPoint placeholders automatically render each line as a bullet. Including bullet characters causes duplication like '- • text'. Use newlines to separate items, and leading spaces (4 spaces) for indentation.
Args: file_path: Path to the document changes: List of changes, each a dict with "target" (cell ref, placeholder, or shape path) and "value" (new content, no bullet chars for PPTX body) track_changes: Log changes for audit trail (default True) output_path: Optional output path (defaults to overwriting input)
Returns: Dictionary with results of all changes. Each successful result includes a "value_preview" field with a truncated preview of the value applied.
Inputs: file_path (required), changes (required), track_changes, output_path, +1 more.
office_readARead-only
Read content from Word, Excel, or PowerPoint documents.
Auto-detects document format from file extension and extracts content in the requested output format.
Replaces: word_extract, word_to_markdown, excel_extract, excel_to_markdown, excel_get_range, pptx_extract, pptx_to_markdown
Examples: # Read entire Excel file as JSON office_read(file_path="data.xlsx")
# Read Excel file as markdown
office_read(file_path="data.xlsx", output_format="markdown")
# Read specific range from Excel
office_read(file_path="data.xlsx", scope="Sheet1!A1:D10")
# Read a single worksheet
office_read(file_path="data.xlsx", scope="Sheet1")
# Read Word document as markdown
office_read(file_path="report.docx", output_format="markdown")
# Read PowerPoint as JSON
office_read(file_path="deck.pptx")Args: file_path: Path to the document (.docx, .xlsx, .xlsm, .pptx) output_format: Output format - "json" for structured data, "markdown" for text representation scope: Optional scope limiter: - Excel: sheet name like "Sheet1" or range like "A1:D10" or "Sheet1!B2:C5" - Word: section title (not yet implemented) - PowerPoint: slide number like "slide:3" (not yet implemented) include_formulas: For Excel, return formulas instead of values
Returns: dict for JSON format, str for markdown format
Inputs: file_path (required), output_format, scope, include_formulas.
office_set_comment_identityA
Set default commenter identity for subsequent comment operations.
This updates in-memory defaults used by office_comment and format-specific add-comment tools when the author argument is omitted.
Args: name: Display name for comments (for example, "Jane Doe") identity: Optional identity string (for example, email or alias) initials: Optional initials override for formats that support it
Returns: Updated identity configuration
Inputs: name (required), identity, initials.
office_tableA
Manage tables in Word, Excel, or PowerPoint documents.
Replaces: excel_get_table, excel_append_table_row, excel_update_table_row, word_get_table, word_insert_table_row, word_patch_table_row, word_create_new_table, pptx_get_table, pptx_insert_table_row, pptx_patch_table_cell
Examples: # Get Excel table data office_table(file_path="data.xlsx", operation="get", table_id="Sales")
# Add row to Excel table
office_table(
file_path="data.xlsx",
operation="add_row",
table_id="Sales",
data={"Product": "Widget", "Amount": 100}
)
# Update Excel table row
office_table(
file_path="data.xlsx",
operation="update_row",
table_id="Sales",
row_index=2,
data={"Amount": 150}
)
# Get Word table (by index, passed as string)
office_table(file_path="report.docx", operation="get", table_id="0")
# Create Word table
office_table(
file_path="report.docx",
operation="create",
data={
"headers": ["Phase", "Owner", "Target Date"],
"rows": [{"Phase": "Discovery", "Owner": "PM", "Target Date": "2026-04-01"}],
"insert_after_section": "Delivery Plan"
}
)
# Get PowerPoint table (slide number as string)
office_table(file_path="deck.pptx", operation="get", table_id="3")Args: file_path: Path to the document operation: "get" to retrieve table data, "add_row" to append a row, "update_row" to modify an existing row, or "create" to create a new table (Word and PowerPoint) table_id: Table identifier as a string. For Excel pass the table name (e.g. "Sales"). For Word pass the 0-based table index for get/add/update (e.g. "0"). For PowerPoint pass the slide number (e.g. "3"). For Word create, table_id is optional. data: Row data as a dict with column names or indices as keys and cell values as values. Required for add_row and update_row. For Word create, provide an object with "headers" and optional "rows", "insert_after_section", "insert_before_section", "output_path", and "author". For PowerPoint update_row include "row", "col", and "value" keys. row_index: 1-based row index for update_row operations
Returns: Dictionary with table data or operation result
Inputs: file_path (required), data (required), operation, table_id, +3 more.
office_templateA
Copy templates or analyze template structure.
Replaces: excel_copy_template, word_copy_template, pptx_copy_template, word_analyze_template_formatting
Examples: # Copy Excel template office_template( source_path="templates/budget.xlsx", destination_path="output/q1-budget.xlsx" )
# Copy Word template
office_template(
source_path="templates/sow.docx",
destination_path="output/acme-sow.docx"
)
# Analyze Word template formatting
office_template(
source_path="templates/sow.docx",
destination_path="", # Not used for analyze
operation="analyze"
)Args: source_path: Path to the template file destination_path: Path for the copy (ignored for analyze) operation: "copy" to copy template, "analyze" to inspect formatting
Returns: Dictionary with operation results
Inputs: source_path (required), destination_path (required), operation.
pptx_add_slideA
Add a new slide to an existing presentation.
Common layout_index values (use pptx_list_masters to see all):
0: Title Slide
1: Title and Content (default - has title + bullet area)
5: Title Only (good for tables or custom content)
6: Blank
Args: file_path: Path to the .pptx file layout_index: Which layout to use (default: 1 = Title and Content) title: Optional title text for the new slide position: Where to insert - 'end' (default), 'start', or 1-based slide number (e.g. '2' to make it slide 2) output_path: Optional output path (defaults to overwriting input)
Returns: Status with new slide_number
Inputs: file_path (required), layout_index, title, position, +1 more.
pptx_add_tableA
Add a data table to a slide.
Creates a table with column headers and optional data rows. Header row is bold. Position/size in inches (16:9 slide = 13.3" × 7.5").
Args: file_path: Path to the .pptx file slide_number: 1-based slide number headers: Column header names (e.g., ['Phase', 'Duration', 'Deliverables']) rows: Data rows as list of lists (optional) left, top: Position from top-left (default: 1.0", 2.0") width, height: Table size (default: 11.0" × 3.0")
Returns: Status with table dimensions
Inputs: file_path (required), slide_number (required), headers (required), rows, +5 more.
pptx_delete_commentADestructive
Delete comments from a slide.
Deletes one comment (by index) or all comments on a slide when comment_index is not provided.
Args: file_path: Path to the .pptx file slide_number: 1-based slide number comment_index: Optional comment index on that slide output_path: Optional output path (defaults to overwriting input)
Returns: Status with deletion details
Inputs: file_path (required), slide_number (required), comment_index, output_path.
pptx_delete_slideADestructive
Remove a slide from the presentation.
USE THIS to remove template slides you don't need or to clean up unwanted content.
Args: file_path: Path to the .pptx file slide_number: 1-based slide number to delete output_path: Optional output path (defaults to overwriting input)
Returns: Status with remaining slide count
Inputs: file_path (required), slide_number (required), output_path.
pptx_duplicate_slideA
Copy a slide including all shapes, tables, and formatting.
USE THIS when you need multiple slides based on a template slide. Performs a full XML deep copy so tables, images, and other shapes are faithfully duplicated.
Args: file_path: Path to the .pptx file slide_number: 1-based slide number to copy position: 'after' (right after original) or 'end' output_path: Optional output path (defaults to overwriting input)
Returns: Status with new slide number
Inputs: file_path (required), slide_number (required), position, output_path.
pptx_from_markdownA
Convert Markdown content to a PowerPoint presentation from inline content or markdown_file.
This is the primary tool for creating PowerPoint decks from text content.
Features:
- Analyzes content to select appropriate layouts for each slide
- Sets theme fonts (default: Segoe UI Semibold for titles, Segoe UI for body)
- Uses millimeters internally for precise positioning
- Fonts inherit from theme - no hardcoded font overrides
Slide Mapping:
- First # heading becomes title slide with large centered title
- Subsequent # or ## headings start new content slides
- --- (horizontal rule) also starts a new slide context
- Bullet points (- or *) become slide body content
- **Label:** patterns are rendered with bold labels (great for key points)
- **Context (assumptions):** after title becomes subtitle on title slide
- Tables (| col | col |) are rendered as PowerPoint tables
- Non-heading, non-bullet paragraphs become plain text
Example:
pptx_from_markdown(
output_path="04. Artifacts/proposal.pptx",
markdown='''
Cloud Migration Proposal
Context: Enterprise transformation for ACME Corp
Executive Summary
Objective: Migrate 15 legacy applications to Azure
Timeline: 12 months with phased approach
Investment
Phase | Cost | Timeline |
Phase 1 | $400K | Q1 |
Phase 2 | $800K | Q2-Q3 |
Next Steps
Approve project charter
Schedule kickoff workshop ''' ) Args: output_path: Path for the output .pptx file markdown: Markdown content following the slide pattern (inline) title_font: Font for titles (default: Segoe UI Semibold) body_font: Font for body text (default: Segoe UI) markdown_file: Optional path to a Markdown file. Use this for very large inputs to avoid MCP argument-size limits. Returns: Status dictionary with file path and slide count
Inputs: output_path (required), markdown, title_font, body_font, +1 more.
pptx_get_notesARead-only
Read speaker notes from slides.
USE THIS to see existing notes before updating them. Without slide_number, returns notes from ALL slides.
Args: file_path: Path to the .pptx file slide_number: Specific slide (omit for all slides)
Returns: Notes content for requested slide(s)
Inputs: file_path (required), slide_number.
pptx_hide_slideA
Hide or unhide a slide in the presentation.
Hidden slides are skipped during slideshow but remain editable. Use for backup content or speaker-only material.
Args: file_path: Path to the .pptx file slide_number: 1-based slide number hidden: True to hide, False to show again output_path: Optional output path (defaults to overwriting input)
Returns: Status with visibility state
Inputs: file_path (required), slide_number (required), hidden, output_path.
pptx_import_slideA
Copy one slide from a source presentation into a target presentation.
USE THIS when you need to lift a fully-designed slide across from one deck to another, preserving pictures, charts, embedded assets, and the layout/master/theme chain required for it to render correctly.
The tool validates the source slide's layout/master chain before the copy. If an identical layout already exists in the target package, it reuses it and does not import a new master. Otherwise it imports only the single layout and single master required by the slide.
Args: source_file_path: Source .pptx file containing the slide to copy source_slide_number: 1-based slide number in the source deck target_file_path: Target .pptx file to receive the slide position: 'end' or 'after' after_slide_number: Required when position='after' output_path: Optional output path (defaults to overwriting target) include_notes: Whether to carry the notes slide across too
Returns: Status dictionary with output path, new slide number, and whether the layout/master were reused or copied.
Inputs: source_file_path (required), source_slide_number (required), target_file_path (required), after_slide_number (required), +3 more.
pptx_list_slidesARead-only
Get an overview of all slides in a PowerPoint file.
USE THIS FIRST when working with an existing presentation to understand its structure before making changes.
Returns for each slide:
number: Slide position (1-based)
title: The slide's title text
layout_name: Which layout template is used
shape_count: Number of shapes on the slide
has_notes: Whether speaker notes exist
hidden: Whether the slide is hidden
Args: file_path: Path to the .pptx file
Returns: Dictionary with slide_count and slides array
Inputs: file_path (required).
pptx_log_changesA
Append change log entries to the first slide's notes.
Use for audit trail since PowerPoint doesn't have track changes. Changes are appended to the notes of slide 1 with timestamps.
Args: file_path: Path to the .pptx file changes: List of change entries, each with slide (number), action (what was done), and detail (specifics) output_path: Optional output path (defaults to overwriting input)
Returns: Status with changes logged count
Inputs: file_path (required), changes (required), output_path.
pptx_recommend_layoutARead-only
Get the best layout for a specific content type.
Content types:
'title': Section title slide
'bullets': Bullet point list
'two_column': Side-by-side content
'comparison': Comparison with headers
'image': Image with caption
'table': Table-heavy content
'blank': Custom content
Args: file_path: Path to the .pptx file content_type: What you want to show on the slide
Returns: Recommended layout_index and alternatives
Inputs: file_path (required), content_type (required).
pptx_reorder_slidesA
Change the order of slides in a presentation.
Provide the complete slide order as a list. For example, [1, 3, 2, 4, 5] moves slide 3 before slide 2.
Args: file_path: Path to the .pptx file new_order: Complete list of slide numbers in desired order (1-based) output_path: Optional output path (defaults to overwriting input)
Returns: Status with new slide order
Inputs: file_path (required), new_order (required), output_path.
pptx_set_notesA
Set speaker notes for a slide.
Speaker notes appear below the slide in Presenter View and can be printed as handouts. Use for talking points and context.
Args: file_path: Path to the .pptx file slide_number: 1-based slide number notes_text: The notes content (supports newlines) append: If True, add to existing notes; if False, replace output_path: Optional output path (defaults to overwriting input)
Returns: Status with notes preview
Inputs: file_path (required), slide_number (required), notes_text (required), append, +1 more.
restart_serverADestructive
Restart the MCP server to reload code changes.
Use this tool after modifying tool modules in .github/mcp/tools/ to pick up the changes without manually restarting.
The server will exit and VS Code will automatically restart it, picking up any code changes in the tools/ directory.
Example: restart_server()
Returns: Status message (the server exits immediately after responding)
Inputs: no named parameters.
web_check_urlARead-only
Check if a URL exists and is accessible.
Performs a HEAD request (or GET if HEAD fails) to verify the URL returns a valid response. Useful for validating links before including them in documents.
Example: web_check_url(url="https://example.com/page")
web_check_url(
url="https://docs.microsoft.com/...",
timeout=5,
follow_redirects=True
)Args: url: The URL to check timeout: Request timeout in seconds (default: 10) follow_redirects: Whether to follow redirects (default: True)
Returns: Dictionary with exists, status_code, final_url, and content_type
Inputs: url (required), timeout, follow_redirects.
web_extract_linksARead-only
Extract all links from a web page.
Fetches a page and extracts all hyperlinks, optionally filtering by pattern or domain. Useful for discovering related pages or building navigation maps.
Example: web_extract_links(url="https://docs.microsoft.com/...")
web_extract_links(
url="https://example.com",
filter_pattern=r"/docs/",
same_domain_only=True
)Args: url: The URL to extract links from filter_pattern: Regex pattern to filter links (optional) same_domain_only: Only return links to the same domain (default: False) timeout: Request timeout in seconds (default: 30)
Returns: Dictionary with extracted links and their text
Inputs: url (required), filter_pattern, same_domain_only, timeout.
web_extract_tablesARead-only
Extract tables from a web page as structured data.
Fetches a page and extracts HTML tables, converting them to a structured format with headers and rows.
Example: web_extract_tables(url="https://example.com/data")
web_extract_tables(
url="https://example.com/report",
table_index=0 # Get only the first table
)Args: url: The URL to extract tables from table_index: Specific table index to extract (optional, 0-based) timeout: Request timeout in seconds (default: 30)
Returns: Dictionary with extracted tables
Inputs: url (required), table_index, timeout.
web_fetchARead-only
Fetch a web page and extract its content as Markdown.
Uses readability to extract the main content from the page, removing navigation, ads, and other clutter. Then converts the clean HTML to Markdown format.
Example: web_fetch(url="https://example.com/article")
web_fetch(
url="https://docs.microsoft.com/en-us/azure/...",
include_links=True,
include_images=True
)Args: url: The URL to fetch extract_content: Use readability to extract main content (default: True) include_links: Include hyperlinks in output (default: True) include_images: Include image references (default: False) timeout: Request timeout in seconds (default: 30)
Returns: Dictionary with title, content (markdown), url, and metadata
Inputs: url (required), extract_content, include_links, include_images, +1 more.
web_searchARead-only
Search the web using DuckDuckGo and return results.
Performs a web search and returns titles, URLs, and snippets for the top results. Does not fetch the full page content - use web_fetch for that.
Example: web_search(query="Microsoft Fabric lakehouse architecture")
web_search(
query="python-pptx table formatting",
max_results=10
)Args: query: Search query string max_results: Maximum number of results to return (default: 5) region: DuckDuckGo region code (default: "wt-wt" for worldwide)
Returns: Dictionary with search results (title, url, snippet)
Inputs: query (required), max_results, region.
word_accept_all_changesA
Accept all tracked insertions/deletions in a Word document.
Removes deletion markup and converts insertion markup into normal document content.
Inputs: file_path (required), output_path.
word_cleanup_sowADestructive
Clean a SOW document by removing all placeholder and instructional content.
Removes:
Highlighted text (turquoise, yellow - template guidance markers)
Colored text (blue, red, purple - instructions)
Bracket placeholders that weren't filled: <...>, [Template Guidance: ...]
Instruction paragraphs containing guidance keywords
This is the final step after generate_sow to ensure the document is presentation-ready with no visible template artifacts. All removals are tracked for auditability.
Example: cleanup_sow( file_path="04. Artifacts/contoso-sow.docx", output_path="04. Artifacts/contoso-sow-final.docx" )
Args: file_path: Path to the SOW document to clean output_path: Path for cleaned output (defaults to overwriting input) author: Author name for tracked changes (default: "Solution Architect Agent")
Returns: Cleanup statistics
Inputs: file_path (required), output_path, author.
word_create_sow_from_markdownA
Create a SOW document from inline markdown or markdown_file by filling a template.
IMPORTANT: This tool requires a template to preserve document structure,
formatting, and corporate styling. It extracts data from Markdown and
uses generate_sow to fill the template.
Workflow:
1. Parse the Markdown to extract structured SOW data
2. Load the template document
3. Fill placeholders and tables with extracted data
4. Save the result
Example:
create_sow_from_markdown(
output_path="04. Artifacts/contoso-sow.docx",
template_path=".github/skills/statement-of-work/templates/Agile.docx",
markdown='''
Contoso – Cloud Migration – Statement of Work
1. Engagement Overview
Customer: Contoso Ltd Provider: Microsoft Project: Cloud Migration Sprint 1
1.1 Business Objectives
Objective | Activities | Assumptions |
Migrate 15 apps | Assessment, planning | Apps are containerizable |
''' |
)
Args:
output_path: Path for the output .docx file
template_path: Path to the .docx template (REQUIRED)
markdown: Markdown content of the SOW (inline)
markdown_file: Optional path to a Markdown file. Use this for
very large inputs to avoid MCP argument-size limits.
Returns:
Status dictionary with file path and extraction summary
Inputs: output_path (required), template_path (required), markdown, markdown_file, +1 more.
word_delete_commentADestructive
Delete a comment from a Word document by comment ID.
Removes the comment from comments.xml and strips associated reference markers from document.xml.
Args: file_path: Path to the .docx file comment_id: Comment ID from word_get_comments output output_path: Optional output path (defaults to overwriting input)
Returns: Status dictionary with deletion details
Inputs: file_path (required), comment_id (required), output_path.
word_document_mapC
Return a lightweight structured document map for a Word file.
Inputs: file_path (required).
word_enable_track_changesA
Enable Track Changes mode in a Word document.
Sets the document settings so that Word will track subsequent changes when the document is opened and edited.
Note: This sets the tracking flag, but changes made by python-docx after this point will NOT be automatically tracked. Use patch_with_track_changes for programmatic tracked edits.
Example: enable_track_changes(file_path="04. Artifacts/contoso-sow.docx")
Args: file_path: Path to the .docx document output_path: Optional output path (defaults to overwriting input)
Returns: Status message
Inputs: file_path (required), output_path.
word_extract_sow_structureARead-only
Extract structured data from an existing SOW document.
Parses a SOW document and extracts key information into a structured format that can be used to generate new documents.
Example: extract_sow_structure( file_path="01. Inputs/existing-sow.docx" )
Args: file_path: Path to the SOW document
Returns: Dictionary with extracted SOW data
Inputs: file_path (required).
word_from_markdownA
Convert Markdown to a Word document from inline content or markdown_file.
This is the primary tool for creating Word documents from text content.
Supports GitHub Flavored Markdown:
- Headings (# ## ### ####)
- Bullet lists (- or *)
- Numbered lists (1. 2. 3.)
- Task lists (- [ ] and - [x])
- Bold (**text**) and italic (*text*) inline formatting
- Strikethrough (~~text~~)
- Inline code (`code`) rendered in Consolas font
- Tables (| col | col |) with proper formatting
- Code blocks (```) rendered in monospace with language hints
- Horizontal rules (---)
Example:
word_from_markdown(
output_path="04. Artifacts/report.docx",
markdown='''
Project Status Report
Executive Summary
The project is on track for Q4 delivery with no minor delays.
Key Metrics
Metric | Value | Status |
Budget | $120,000 | On track |
Timeline | Q4 2026 | Green |
Quality | 95% | Exceeds |
Next Steps
Complete UAT testing
Finalize documentation
Schedule go-live review
Phase 1 complete
Phase 2 in progress
Phase 3 planned ''' ) Args: output_path: Path for the output .docx file markdown: Full GitHub Flavored Markdown content (inline) markdown_file: Optional path to a Markdown file. Use this for very large documents to avoid MCP argument-size limits. Returns: Status dictionary with file path
Inputs: output_path (required), markdown, markdown_file.
word_generate_sowA
Generate a SOW document from a template and structured data.
IMPORTANT: This tool REQUIRES a template document. It fills placeholders and tables but does NOT generate prose sections (Introduction, Executive Summary, etc.). After using this tool, use patch_section to add narrative content to key sections.
Template-based workflow:
Use copy_template to copy .github/skills/statement-of-work/templates/Agile.docx
Use generate_sow to fill placeholders and tables
Use patch_section to add Introduction, Business Context, etc.
Use audit_completion to verify completeness
Use cleanup_sow to remove template artifacts
Takes a SOW template and fills it with actual engagement data, stripping instructional boilerplate and replacing placeholders.
Example: generate_sow( template_path=".github/skills/statement-of-work/templates/Agile.docx", output_path="04. Artifacts/contoso-sow.docx", sow_data={ "customer_name": "Contoso", "customer_short_name": "Contoso", "project_name": "Cloud Migration", "provider_name": "Microsoft", "work_order_number": "WO-2026-001", "language": "English", "business_objectives": [ {"objective": "Migrate 15 apps to Azure", "activities": "Assessment, migration planning", "assumptions": "Apps are containerizable"} ], "epics": [ {"name": "Infrastructure Setup", "description": "Set up Azure landing zone", "assumptions": "Subscription available"} ], "out_of_scope": [ {"area": "Data migration", "description": "Historical data migration not included"} ], "technology_requirements": [ {"item": "Azure subscription", "version": "N/A", "ready_by": "Project start"} ], "assumptions": [ "Customer will provide access to existing systems", "Dedicated product owner available full-time" ] } )
Args: template_path: Path to the .docx template (REQUIRED - use copy_template first) output_path: Path for the output .docx file sow_data: Dictionary containing SOW content
Returns: Status dictionary with file path and next_tools suggestions
Inputs: template_path (required), output_path (required), sow_data (required), mode.
word_get_commentsARead-only
Extract comments from a Word document with metadata and thread context.
Args: file_path: Path to the .docx file filter: Optional filter (all/open/resolved/mine) author: Author name for filter='mine' (defaults to comment identity) format: flat (default) or threaded
Returns: Dictionary with comment list/count and optional thread groups
Inputs: file_path (required), filter, author, format.
word_get_section_guidanceARead-only
Extract template guidance and instructions from a section.
Reads a section and identifies all instructional content that tells you what to write there. This includes:
Template guidance markers: [Template Guidance: ...]
Highlighted text (turquoise, yellow) indicating placeholders
Colored text (blue, red, purple) indicating instructions
Insert instructions: [insert ...]
Use this BEFORE patch_section to understand what content is expected.
Example: get_section_guidance( file_path="04. Artifacts/contoso-sow.docx", section_title="Introduction" )
Args: file_path: Path to the .docx document section_title: Title or partial title of the section
Returns: Structured guidance including instructions, expected content hints, and placeholders to fill
Inputs: file_path (required), section_title (required).
word_insert_at_anchorB
Insert paragraphs before/after a matched anchor or paragraph index.
This is a general-purpose insertion tool for narrative content when a document has a stable anchor paragraph but you do not want to replace the entire section body.
Inputs: file_path (required), content (required), anchor_text, paragraph_index, +4 more.
word_list_anchorsARead-only
List likely anchor paragraphs and headings for insertion workflows.
Returns headings plus high-signal non-empty paragraphs that can be used
with word_insert_at_anchor.
Inputs: file_path (required), query, include_paragraphs.
word_parse_sow_templateC
Parse a SOW template to extract its structure.
Analyzes a Word document template to identify:
Document sections and headings
Tables and their purposes
Placeholder variables that need filling
Instructional text to be stripped
Example: parse_sow_template( template_path=".github/skills/statement-of-work/templates/Agile.docx" )
Args: template_path: Path to the .docx template file
Returns: Dictionary with template structure analysis
Inputs: template_path (required).
word_patch_with_track_changesA
Replace text in a document with Word Track Changes enabled.
Creates revision marks (insertions/deletions) that appear in Word's review mode. Old text is marked as deleted (red strikethrough) and new text is marked as inserted (green underline).
This is useful for:
Auditable document changes
Review workflows where changes need approval
Comparing before/after states in Word
Example: patch_with_track_changes( file_path="04. Artifacts/contoso-sow.docx", replacements={ "": "Contoso Ltd", "": "Cloud Migration", "[TBD]": "Q1 2026" }, author="Solution Architect" )
Args: file_path: Path to the .docx document replacements: Dictionary mapping old text to new text author: Name to attribute changes to (appears in Word's review pane) output_path: Optional output path (defaults to overwriting input)
Returns: Status with replacement counts
Inputs: file_path (required), replacements (required), author, output_path.
word_reply_commentC
Backward-compatible alias for word_reply_to_comment.
Inputs: file_path (required), comment_id (required), reply_text (required), author, +2 more.
word_reply_to_commentA
Add a threaded reply to an existing Word comment.
Creates a new w:comment entry in word/comments.xml and links it
to the parent thread using w14:paraIdParent on the reply's first
paragraph. If the parent comment has no w14:paraId (older docs), a
synthetic one is added and reused.
Args: file_path: Path to the .docx file comment_id: ID of the parent comment (from word_get_comments) text: Reply text author: Reply author (defaults to office_set_comment_identity/env) output_path: Optional output path (defaults to overwriting input) auto_resolve: Mark the thread resolved after adding the reply
Returns: Status dictionary with reply details
Inputs: file_path (required), comment_id (required), text (required), author, +2 more.
word_resolve_commentA
Mark a Word comment thread as resolved or open.
Resolution state is stored in word/commentsExtended.xml (w15:commentEx@w15:done). If a reply comment ID is supplied, the root thread comment is updated.
Args: file_path: Path to the .docx file comment_id: Comment ID from word_get_comments resolved: True to resolve, False to reopen output_path: Optional output path (defaults to overwriting input)
Returns: Status dictionary with thread resolution details
Inputs: file_path (required), comment_id (required), resolved (required), output_path.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 61 tools
Many tools overlap, e.g., office_inspect duplicates excel_list_sheets, word_list_sections, etc., and office_read replaces multiple extraction tools. Also, word_create_sow_from_markdown and word_generate_sow have similar descriptions, causing potential confusion.
Tool names predominantly follow a snake_case verb_noun pattern (e.g., azure_calculate_cost, excel_add_chart). Minor deviations include `office_help` (not a specific action) and `restart_server`, but overall the pattern is consistent.
With 61 tools, the server is overburdened for its office-document focus. The inclusion of Azure pricing and web utilities bloats the surface. Many tools could be consolidated (e.g., unified office_* tools already exist alongside format-specific ones).
Office document capabilities are thorough: create, read, update, delete, and inspection for Word, Excel, and PowerPoint. Minor gaps exist (e.g., no PowerPoint chart creation), but the core workflows are covered. Azure and web utilities add extra but incomplete coverage.
Maintenance
Related MCP Connectors
Convert, create and edit Word, PowerPoint, Excel and PDF documents from any AI agent.
Create Word .docx files from sections or markdown, fill templates, and read text from a .docx.
Generate PDF, Word (.docx) and PowerPoint (.pptx) documents from Markdown over MCP.
Generate, edit, merge, translate and PDF-convert PowerPoint (.pptx) over MCP. 8 tools.
Related MCP Servers
- FlicenseAqualityDmaintenanceA comprehensive Model Context Protocol server that processes Microsoft Word documents with full formatting support, enabling text extraction, HTML/Markdown conversion, structure analysis, and image extraction.52-
- AlicenseNot gradedqualityDmaintenanceEnables comprehensive Microsoft Word document manipulation through the Model Context Protocol, with advanced table operations including creation, data management, formatting, and bulk operations. Supports document creation, editing, and saving with plans for full document content management.7MIT
- AlicenseDqualityDmaintenanceEnables reading, writing, editing, and converting Office documents (ODT, DOCX, ODS, XLSX, PDF, etc.) using MCP tools, with no external dependencies.1121 npmMIT
- AlicenseCqualityDmaintenanceA Model Context Protocol server for enterprise-grade document automation, enabling AI assistants to create, read, manipulate, and analyze Microsoft Word documents programmatically.541MIT