// MCP TOKEN COST

The Servicialo MCP server costs 18,603 tokens before the first call.

Every request your agent makes carries every tool definition this server exposes — context your code, documents and conversation can't use, mostly for tools the agent never calls. You don't need them all in the window, and you don't have to pay for them.

QUICK ANSWER The Servicialo MCP server's 111 tool definitions consume 18,603 tokens — 9.3% of a 200k context window, and 9.1× the median MCP server (2,038 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS tiktoken o200k_base · rank #148 of 6,689 measured servers · refreshed every build Method →

// CONTEXT WINDOW SHARE

What that costs 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 9.3%

1M WINDOW 1.9%

Corpus context: Servicialo ranks #148 of 6,689 measured MCP servers by definition cost. The median is 2,038 tokens, p90 is 12,122, and the heaviest (Ainumbers Mcp Apps) is 269,012 — 135% of a 200k window on its own. New to this? See MCP token cost and context window in the glossary.

Token cost is one axis. See the risk picture across your whole stack →

// PER-TOOL BREAKDOWN

Where the 18,603 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 168 tokens per tool.

Tool	Category	Tokens	% of server
reminders_update_config	Write	561	3.0%
public_booking_create	Write	497	2.7%
agendas_update	Write	409	2.2%
booking_create	Write	387	2.1%
comms_create_campaign	Write	371	2.0%
scheduling_book	Write	365	2.0%
service_create	Write	341	1.8%
lifecycle_history	Read	340	1.8%
agendas_create	Write	328	1.8%
booking_create_recurring	Write	322	1.7%
session_note_upsert	Write	298	1.6%
lifecycle_transition	Execute	296	1.6%
service_update	Write	295	1.6%
availability_get_slots	Read	281	1.5%
provider_create	Write	246	1.3%
client_create	Write	239	1.3%
delivery_confirmations_list	Read	233	1.3%
settings_get	Read	228	1.2%
comms_render_message	Write	227	1.2%
booking_update_status	Write	220	1.2%
booking_list_requests	Read	217	1.2%
finance_list_gastos	Read	214	1.2%
admin_create_service	Write	211	1.1%
portal_confirm_delivery	Read	198	1.1%
scheduling_cancel	Write	195	1.0%
portal_propose_reschedule	Write	193	1.0%
admin_toggle_discoverable	Write	188	1.0%
dunning_configure	Write	188	1.0%
finance_create_venta	Financial	187	1.0%
admin_set_availability	Write	185	1.0%
booking_resolve_request	Write	184	1.0%
booking_list	Read	182	1.0%
portal_report_session	Write	181	1.0%
treatment_plans_list	Read	176	0.9%
members_invite	Write	176	0.9%
finance_register_payment	Financial	175	0.9%
comms_update_preferences	Write	174	0.9%
booking_cancel	Financial	173	0.9%
lifecycle_get_state	Read	173	0.9%
portal_session_cancel	Destructive	171	0.9%
portal_session_confirm	Write	171	0.9%
comms_send_message	Write	168	0.9%
settings_update	Write	165	0.9%
finance_list_invoices	Read	163	0.9%
provider_update	Write	159	0.9%
org_update	Write	155	0.8%
portal_cancel_reschedule_request	Destructive	153	0.8%
reminders_get_config	Read	152	0.8%
session_note_get	Read	152	0.8%
finance_create_cobro	Financial	151	0.8%
finance_list_cobros	Read	149	0.8%
comms_send_campaign	Execute	149	0.8%
finance_list_ventas	Read	146	0.8%
service_assign_provider	Write	144	0.8%
comms_list_audiences	Read	141	0.8%
client_update	Write	141	0.8%
public_availability_get_slots	Read	139	0.7%
resources_list	Read	138	0.7%
finance_send_confirmations	Financial	133	0.7%
booking_reschedule	Write	132	0.7%
cierre_distribuir_utilidades	Financial	132	0.7%
public_booking_reschedule	Write	132	0.7%
scheduling_reschedule	Write	131	0.7%
scheduling_confirm	Write	128	0.7%
cierre_crear_cliente	Financial	127	0.7%
finance_aging	Read	124	0.7%
report_no_shows	Read	124	0.7%
disputes_list	Read	123	0.7%
finance_list_confirmations	Read	123	0.7%
org_summary	Read	122	0.7%
public_booking_cancel	Destructive	120	0.6%
finance_list_payments	Read	120	0.6%
report_deuda_real	Read	120	0.6%
report_sc_summary	Read	119	0.6%
treatment_plans_get	Read	118	0.6%
agendas_get	Read	117	0.6%
client_list	Read	116	0.6%
email_domain_register	Write	115	0.6%
comms_list_campaigns	Read	114	0.6%
report_occupancy	Read	114	0.6%
public_booking_get	Read	111	0.6%
report_revenue	Read	110	0.6%
payroll_list_records	Read	109	0.6%
agendas_delete	Destructive	106	0.6%
cierre_cerrar_org	Destructive	106	0.6%
public_booking_confirm	Write	106	0.6%
cierre_preview_cliente	Read	105	0.6%
email_domain_delete	Destructive	104	0.6%
provider_get_stats	Read	104	0.6%
cierre_eliminar_cliente	Destructive	103	0.6%
comms_get_campaign	Read	103	0.6%
admin_list_providers	Read	101	0.5%
cierre_evaluar_org	Read	100	0.5%
email_domain_verify	Execute	100	0.5%
service_list	Read	96	0.5%
client_get	Read	95	0.5%
public_service_list	Read	95	0.5%
availability_get_provider_schedule	Read	94	0.5%
cierre_listar_utilidades	Read	93	0.5%
email_domain_get	Read	93	0.5%
booking_get	Read	91	0.5%
cierre_listar_clientes	Read	89	0.5%
finance_client_balance	Read	88	0.5%
finance_get_cobro	Read	88	0.5%
dunning_get_config	Read	87	0.5%
payroll_get_summary	Read	84	0.5%
report_dashboard	Read	81	0.4%
provider_get	Read	80	0.4%
comms_get_preferences	Read	77	0.4%
agendas_list	Read	74	0.4%
members_list	Read	65	0.3%

// CUT THE COST

Your agent uses a handful of these tools. It pays for all 111.

You don't need all 111 of those definitions in the window. PolicyLayer is an MCP gateway that sits in front of Servicialo: only the tools you grant are exposed to the agent, the rest never load. A smaller window means a sharper agent — less noise when it picks a tool — and every request costs less:

Grant scope	Definition cost	Reduction
All 111 tools (no gateway)	18,603 tokens	—
3 granted tools	~503 tokens	−97%
5 granted tools	~838 tokens	−95%
10 granted tools	~1,676 tokens	−91%

The risk dividend: 14 of these 111 tools are critical-risk (destructive or financial) and cost 1,941 tokens (10% of the definition load). Block them — the recommended starter policy — and you reclaim that context before tuning anything else.

Create a free account and register Servicialo — nothing to install.
Grant only the tools you use — ungranted definitions never enter the context window.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

CUT SERVICIALO TOKEN COST →

Instant setup, no code required.

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

//FAQ

Servicialo token-cost questions.

How many tokens does the Servicialo MCP server use?+

Its 111 tool definitions total 18,603 tokens — 9.3% 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 Servicialo 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 Servicialo's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Servicialo 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 503 tokens, a 97% 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 17-08-2026 from the PolicyLayer scan database over all 111 catalogued Servicialo tools. Counts refresh with every site build.

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

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

CUT SERVICIALO TOKEN COST →

Instant setup, no code required.

46,500+ MCP servers and 515,000+ tools scanned and risk-classified.