// MCP TOKEN COST

The Excel Webview2 MCP server costs 6,996 tokens before the first call.

Connect Excel Webview2 and its 59 tool definitions are loaded into the model's context on every request — 3.5% of a 200k window spent before your agent does anything.

QUICK ANSWER The Excel Webview2 MCP server's tool definitions consume 6,996 tokens — 3.7× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 59 tools · 6,996 tokens · 3.5% of 200k · 0.7% of 1M Method →

// CONTEXT WINDOW SHARE

What that buys before your agent starts working.

Tool definitions are overhead: they occupy context on every request and compete with your code, documents and conversation history for the same window.

200K WINDOW 3.5%

1M WINDOW 0.7%

Corpus context: Excel Webview2 ranks #903 of 3,213 measured MCP servers by definition cost. The median is 1,905 tokens, p90 is 7,952, and the heaviest (Fusionauth) is 183,337 — 92% of a 200k window on its own.

// PER-TOOL BREAKDOWN

Where the 6,996 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 119 tokens per tool.

Tool	Category	Tokens	% of server
take_screenshot	Read	233	3.3%
list_network_requests	Read	232	3.3%
list_console_messages	Read	230	3.3%
excel_range_special_cells	Read	225	3.2%
excel_read_range	Read	202	2.9%
evaluate_script	Read	197	2.8%
excel_find_in_range	Read	192	2.7%
excel_range_properties	Read	192	2.7%
performance_start_trace	Execute	180	2.6%
take_snapshot	Read	167	2.4%
excel_used_range	Read	161	2.3%
excel_launch_addin	Execute	160	2.3%
get_network_request	Read	159	2.3%
press_key	Write	149	2.1%
performance_analyze_insight	Execute	148	2.1%
lighthouse_audit	Execute	147	2.1%
fill_form	Write	142	2.0%
emulate	Execute	137	2.0%
excel_list_conditional_formats	Read	137	2.0%
excel_chart_image	Read	136	1.9%
excel_list_data_validations	Read	134	1.9%
excel_range_formulas	Read	130	1.9%
upload_file	Write	129	1.8%
fill	Write	124	1.8%
excel_active_range	Execute	121	1.7%
wait_for	Execute	120	1.7%
click	Write	115	1.6%
drag	Write	113	1.6%
select_page	Write	104	1.5%
excel_table_rows	Read	101	1.4%
handle_dialog	Write	100	1.4%
performance_stop_trace	Execute	99	1.4%
excel_detect_addin	Read	96	1.4%
take_memory_snapshot	Read	96	1.4%
type_text	Write	96	1.4%
excel_chart_info	Read	95	1.4%
hover	Read	94	1.3%
excel_table_info	Read	92	1.3%
connection_status	Read	90	1.3%
get_console_message	Read	89	1.3%
excel_stop_addin	Execute	85	1.2%
excel_settings_get	Read	85	1.2%
close_page	Write	85	1.2%
excel_pivot_table_info	Read	84	1.2%
excel_worksheet_info	Read	84	1.2%
excel_list_charts	Read	82	1.2%
excel_list_shapes	Read	81	1.2%
excel_table_filters	Read	80	1.1%
excel_pivot_table_values	Read	79	1.1%
excel_list_comments	Read	77	1.1%
excel_list_tables	Read	65	0.9%
excel_list_named_items	Read	63	0.9%
excel_calculation_state	Read	60	0.9%
excel_list_pivot_tables	Read	57	0.8%
excel_list_worksheets	Read	57	0.8%
excel_workbook_info	Write	56	0.8%
excel_context_info	Read	55	0.8%
excel_custom_xml_parts	Read	51	0.7%
list_pages	Read	46	0.7%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 59.

A PolicyLayer grant exposes only the tools you allow — ungranted definitions are filtered out of the tool list, so they never enter the context window. Estimates below assume typical-weight tools (119 tokens each).

Grant scope	Definition cost	Reduction
All 59 tools (no gateway)	6,996 tokens	—
3 granted tools	~356 tokens	−95%
5 granted tools	~593 tokens	−92%
10 granted tools	~1,186 tokens	−83%

Model your own stack in the token-cost calculator, or see the Excel Webview2 policy for what a sensible grant looks like.

//FAQ

Excel Webview2 token-cost questions.

How many tokens does the Excel Webview2 MCP server use?+

Its 59 tool definitions total 6,996 tokens — 3.5% of a 200k context window — measured with tiktoken o200k_base over the serialised tools/list payload. Exact counts vary slightly by client and model.

Why does Excel Webview2 consume tokens before I send a message?+

MCP clients load every connected server's tool definitions — name, description, and input schema — into the model's context so it knows what it can call. That payload is charged against your context window on every request, whether or not a tool is used.

How do I reduce Excel Webview2's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Excel Webview2 to only the tools you allow — ungranted definitions are filtered out of the tool list, so they never enter the context window. A grant of 3 typical tools costs roughly 356 tokens, a 95% reduction.

Does deferred tool loading fix this?+

Partially, in some clients. Claude Code defers MCP tool schemas behind a tool-search step by default, and VS Code has experimental grouping — but you still pay tokens per search and reload, and Cursor, Windsurf and Gemini CLI load definitions upfront. Reducing the exposed tool set cuts the cost in every client.

// METHOD & REVIEW

How these numbers were measured.

Serialisation

Each tool is serialised as a tools/list entry — name, description, input schema — from the schemas in the PolicyLayer scan database. Clients differ slightly in framing, so treat counts as close estimates.

Tokeniser

tiktoken o200k_base (GPT-4o/o-series). Anthropic's current tokeniser isn't published, so Claude's exact counts will differ; for English text and JSON schemas the totals are close enough to treat these as estimates.

Deferred loading

Some clients now defer schema loading (Claude Code's tool search; VS Code experimental grouping). You still pay per search and reload — and Cursor, Windsurf and Gemini CLI load everything upfront.

Computed 07-06-2026 from the PolicyLayer scan database over all 59 catalogued Excel Webview2 tools. Counts refresh with every site build.

Expose only the tools you use — the rest never enter your context.

A PolicyLayer grant scopes Excel Webview2 to the tools you actually allow. Ungranted definitions never load, and every call that does run is checked against policy first.

SCOPE YOUR GRANT →

Free to start. No card required.

4,600+ MCP servers and 31,000+ tools scanned and risk-classified.