// MCP TOKEN COST

The Screenshots MCP server costs 7,509 tokens before the first call.

Connect Screenshots and its 56 tool definitions are loaded into the model's context on every request — 3.8% of a 200k window spent before your agent does anything.

QUICK ANSWER The Screenshotsmcp MCP server's tool definitions consume 7,509 tokens — 3.9× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 56 tools · 7,509 tokens · 3.8% of 200k · 0.8% 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 3.8%

1M WINDOW 0.8%

Corpus context: Screenshots ranks #675 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 7,509 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 134 tokens per tool.

Tool	Category	Tokens	% of server
browser_navigate	Execute	521	6.9%
auth_test_assist	Read	407	5.4%
take_screenshot	Read	267	3.6%
smart_login	Write	245	3.3%
solve_captcha	Read	225	3.0%
browser_cookies	Destructive	221	2.9%
screenshot_element	Read	191	2.5%
screenshot_batch	Read	190	2.5%
browser_click_at	Write	186	2.5%
screenshot_diff	Read	185	2.5%
browser_storage	Write	176	2.3%
create_test_inbox	Write	173	2.3%
browser_network_requests	Read	169	2.3%
browser_get_accessibility_tree	Read	166	2.2%
read_verification_email	Read	165	2.2%
accessibility_snapshot	Read	157	2.1%
screenshot_fullpage	Read	154	2.1%
screenshot_responsive	Read	154	2.1%
screenshot_dark	Read	149	2.0%
ux_review	Execute	146	1.9%
screenshot_tablet	Read	144	1.9%
webhook_create	Write	144	1.9%
screenshot_cross_browser	Read	138	1.8%
og_preview	Read	136	1.8%
browser_set_viewport	Write	136	1.8%
browser_get_html	Read	127	1.7%
browser_console_logs	Read	122	1.6%
check_inbox	Read	119	1.6%
browser_press_key	Write	110	1.5%
send_test_email	Write	105	1.4%
browser_fill	Write	103	1.4%
browser_wait_for	Execute	101	1.3%
browser_evaluate	Execute	98	1.3%
webhook_deliveries	Read	97	1.3%
browser_network_errors	Read	95	1.3%
browser_click	Write	95	1.3%
find_login_page	Read	93	1.2%
browser_get_text	Read	92	1.2%
browser_select_option	Write	90	1.2%
browser_scroll	Write	88	1.2%
find_breakpoints	Read	82	1.1%
browser_hover	Read	79	1.1%
browser_seo_audit	Read	79	1.1%
browser_perf_metrics	Read	77	1.0%
webhook_test	Read	73	1.0%
browser_close	Write	69	0.9%
authorize_email_access	Read	67	0.9%
list_recent_screenshots	Read	63	0.8%
get_screenshot_status	Read	62	0.8%
screenshot_pdf	Read	62	0.8%
webhook_rotate	Write	58	0.8%
webhook_delete	Destructive	56	0.7%
browser_go_back	Read	55	0.7%
webhook_list	Read	54	0.7%
browser_screenshot	Read	48	0.6%
browser_go_forward	Read	45	0.6%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 56.

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 (134 tokens each).

Grant scope	Definition cost	Reduction
All 56 tools (no gateway)	7,509 tokens	—
3 granted tools	~402 tokens	−95%
5 granted tools	~670 tokens	−91%
10 granted tools	~1,341 tokens	−82%

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

//FAQ

Screenshotsmcp token-cost questions.

How many tokens does the Screenshots MCP server use?+

Its 56 tool definitions total 7,509 tokens — 3.8% 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 Screenshotsmcp 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 Screenshotsmcp's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Screenshotsmcp 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 402 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 56 catalogued Screenshotsmcp tools. Counts refresh with every site build.

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

A PolicyLayer grant scopes Screenshotsmcp 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.