// MCP TOKEN COST

The Looking Glass MCP server costs 5,080 tokens before the first call.

Connect Looking Glass and its 71 tool definitions are loaded into the model's context on every request — 2.5% of a 200k window spent before your agent does anything.

QUICK ANSWER The Looking Glass MCP server's tool definitions consume 5,080 tokens — 2.7× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 71 tools · 5,080 tokens · 2.5% of 200k · 0.5% 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 2.5%

1M WINDOW 0.5%

Corpus context: Looking Glass ranks #1023 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 5,080 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 72 tokens per tool.

Tool	Category	Tokens	% of server
test_scenario_run	Read	520	10.2%
test_auth_flow	Read	182	3.6%
browser_login	Write	150	3.0%
test_assert	Read	135	2.7%
test_fill_form	Read	134	2.6%
test_chaos	Read	130	2.6%
test_watch_events	Read	127	2.5%
browser_mock_route	Read	120	2.4%
browser_extract	Read	117	2.3%
browser_explore	Read	103	2.0%
browser_set_cookie	Write	96	1.9%
browser_screenshot	Read	95	1.9%
browser_wait_for	Execute	94	1.9%
browser_fill_and_submit	Write	94	1.9%
browser_scroll	Write	94	1.9%
browser_network_requests	Read	86	1.7%
session_export_playwright	Write	86	1.7%
browser_wait_until_stable	Execute	82	1.6%
browser_diagnose	Read	78	1.5%
browser_drag	Write	76	1.5%
vault_login	Write	75	1.5%
browser_type	Write	74	1.5%
browser_smart_click	Write	73	1.4%
browser_console_messages	Read	72	1.4%
browser_resolve_element	Write	72	1.4%
browser_select_option	Write	72	1.4%
browser_go	Read	69	1.4%
browser_set_localstorage	Write	68	1.3%
browser_clear_storage	Destructive	67	1.3%
browser_press_key	Write	67	1.3%
browser_tab_select	Write	67	1.3%
session_replay	Read	64	1.3%
visual_baseline	Read	64	1.3%
vault_inject	Read	62	1.2%
browser_evaluate	Read	60	1.2%
browser_action_history	Read	57	1.1%
browser_get_localstorage	Read	57	1.1%
browser_click	Write	56	1.1%
browser_close	Write	56	1.1%
browser_navigate	Execute	55	1.1%
browser_hover	Read	55	1.1%
visual_compare	Read	55	1.1%
vault_delete	Destructive	54	1.1%
browser_tab_new	Execute	52	1.0%
browser_get_cookies	Read	51	1.0%
session_start	Execute	48	0.9%
browser_performance_audit	Execute	36	0.7%
test_accessibility_audit	Read	36	0.7%
test_chaos_clear	Read	36	0.7%
browser_clear_mocks	Destructive	35	0.7%
browser_analyze_page	Read	35	0.7%
browser_clean_slate	Read	35	0.7%
browser_list_mocks	Read	35	0.7%
browser_suggest_actions	Read	35	0.7%
test_generate_assertions	Read	35	0.7%
test_scenario_status	Read	35	0.7%
browser_clear_history	Destructive	34	0.7%
browser_diff_state	Read	34	0.7%
browser_error_report	Read	34	0.7%
browser_go_back	Read	34	0.7%
browser_go_forward	Read	34	0.7%
browser_health_check	Read	34	0.7%
browser_recover	Read	34	0.7%
browser_snapshot_state	Read	34	0.7%
browser_tab_list	Read	34	0.7%
browser_workflow	Read	34	0.7%
test_stop_watch	Read	34	0.7%
browser_snapshot	Read	33	0.6%
session_end	Read	33	0.6%
session_list	Read	33	0.6%
vault_list	Read	33	0.6%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 71.

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 (72 tokens each).

Grant scope	Definition cost	Reduction
All 71 tools (no gateway)	5,080 tokens	—
3 granted tools	~215 tokens	−96%
5 granted tools	~358 tokens	−93%
10 granted tools	~715 tokens	−86%

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

//FAQ

Looking Glass token-cost questions.

How many tokens does the Looking Glass MCP server use?+

Its 71 tool definitions total 5,080 tokens — 2.5% 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 Looking Glass 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 Looking Glass's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Looking Glass 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 215 tokens, a 96% 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.

01

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.

02

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.

03

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 71 catalogued Looking Glass tools. Counts refresh with every site build.