// MCP TOKEN COST

The Mcp Singapore MCP server costs 26,187 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 Mcp Singapore MCP server's 291 tool definitions consume 26,187 tokens — 13% of a 200k context window, and 14× the median MCP server (1,866 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS tiktoken o200k_base · rank #231 of 12,801 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 13%

1M WINDOW 2.6%

Corpus context: Mcp Singapore ranks #231 of 12,801 measured MCP servers by definition cost. The median is 1,866 tokens, p90 is 13,646, and the heaviest (Ainumbers Mcp Apps) is 442,891 — 221% 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 26,187 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 90 tokens per tool.

Tool	Category	Tokens	% of server
insights_compare_series	Read	310	1.2%
sg_cross_dataset	Read	310	1.2%
hdb_resale_search	Read	285	1.1%
sg_hdb_resale_stats	Read	283	1.1%
hdb_resale_stats	Read	282	1.1%
insights_align_series	Read	217	0.8%
sg_ecda_vacancy_summary	Read	211	0.8%
ecda_vacancy_summary	Read	210	0.8%
sg_moe_schools_near	Read	204	0.8%
moe_schools_near	Read	203	0.8%
sg_ecda_centres_near	Read	202	0.8%
ecda_centres_near	Read	201	0.8%
finance_mortgage_affordability	Read	195	0.7%
rail_list_stations	Read	171	0.7%
datagov_get_realtime	Read	167	0.6%
rail_nearest_stations	Read	164	0.6%
sg_search_datasets	Read	164	0.6%
singapore_dataset_search	Read	164	0.6%
onemap_route	Read	162	0.6%
insights_visualize_series	Write	162	0.6%
sg_fx_basket	Write	154	0.6%
rail_nearest_stations_to_address	Read	153	0.6%
fx_basket	Write	153	0.6%
sg_visualize	Read	152	0.6%
bank_interest_rates_search	Read	140	0.5%
rail_nearest_exits	Read	137	0.5%
ura_private_property_search	Read	134	0.5%
coe_bidding_search	Read	130	0.5%
ecda_childcare_search	Read	130	0.5%
hawker_centres_search	Read	130	0.5%
moe_schools_search	Read	130	0.5%
gdp_growth_search	Read	129	0.5%
hdb_carparks_search	Read	129	0.5%
rail_list_exits	Read	129	0.5%
visitor_arrivals_search	Read	129	0.5%
dengue_clusters_search	Read	128	0.5%
disease_cases_search	Read	128	0.5%
median_income_search	Read	127	0.5%
electricity_search	Read	126	0.5%
finance_mortgage_payment	Read	126	0.5%
iras_tax_search	Read	126	0.5%
population_search	Read	126	0.5%
retail_sales_search	Read	126	0.5%
cpi_search	Read	125	0.5%
employment_sector_search	Read	125	0.5%
crime_search	Read	124	0.5%
live_births_search	Read	124	0.5%
mas_fx_search	Read	124	0.5%
unemployment_search	Read	124	0.5%
singapore_prompt_examples	Read	123	0.5%
tourism_receipts_search	Read	123	0.5%
finance_mortgage_stress_test	Read	122	0.5%
datagov_list_dataset_rows	Read	121	0.5%
rail_list_stations_by_line	Read	120	0.5%
business_formations_compare	Read	117	0.4%
sg_formations_compare	Read	117	0.4%
acra_search_entities	Read	114	0.4%
disease_cases_trend	Read	114	0.4%
onemap_reverse_geocode	Read	114	0.4%
sg_disease_trend	Read	114	0.4%
rail_list_interchanges	Read	113	0.4%
singapore_property_area_brief	Read	113	0.4%
weather_twenty_four_hour_forecast	Read	112	0.4%
rail_historical_station_counts	Read	110	0.4%
onemap_search	Read	109	0.4%
analytics_correlate_series	Read	108	0.4%
weather_four_day_outlook	Read	108	0.4%
weather_two_hour_forecast	Read	108	0.4%
sg_visitors_top_sources	Read	107	0.4%
visitors_top_sources	Read	106	0.4%
weather_air_quality	Read	106	0.4%
weather_rainfall	Read	106	0.4%
weather_temperature	Read	104	0.4%
analytics_compare_series	Read	101	0.4%
sg_coe_demand_supply	Read	101	0.4%
singapore_tool_profiles	Read	101	0.4%
finance_mortgage_rates_history	Read	100	0.4%
coe_demand_supply	Read	99	0.4%
lta_bus_arrivals	Read	99	0.4%
rail_search_stations	Read	97	0.4%
sg_coe_history	Read	96	0.4%
sg_cpi_yoy	Read	96	0.4%
singapore_company_brief	Read	96	0.4%
cpi_yoy	Read	95	0.4%
sg_fx_rate	Read	95	0.4%
sg_retail_sales_yoy	Read	95	0.4%
coe_history	Read	94	0.4%
fx_rate	Read	94	0.4%
retail_sales_yoy	Read	94	0.4%
business_formations_net	Read	93	0.4%
sg_net_formations	Read	93	0.4%
singapore_capability_registry	Read	93	0.4%
bank_interest_rates_metadata	Read	92	0.4%
sg_employment_growth	Read	91	0.3%
sg_fx_history	Read	90	0.3%
singapore_prompt_for_tool	Read	90	0.3%
employment_growth	Read	89	0.3%
fx_history	Read	89	0.3%
rail_get_station_connections	Read	89	0.3%
finance_mortgage_rates_latest	Read	88	0.3%
hdb_resale_metadata	Read	86	0.3%
ura_private_property_metadata	Read	86	0.3%
disease_cases_latest	Read	85	0.3%
sg_disease_latest	Read	85	0.3%
rail_list_lines	Read	84	0.3%
sg_gdp_industry_compare	Read	84	0.3%
singstat_business_cessations_monthly_history	Read	84	0.3%
gdp_industry_compare	Read	83	0.3%
singstat_business_formations_annual_history	Read	83	0.3%
singstat_business_formations_monthly_history	Read	83	0.3%
coe_bidding_metadata	Read	82	0.3%
datagov_get_dataset_metadata	Read	82	0.3%
ecda_childcare_metadata	Read	82	0.3%
hawker_centres_metadata	Read	82	0.3%
moe_schools_metadata	Read	82	0.3%
singstat_merchandise_trade_history	Read	82	0.3%
singstat_labour_force_history	Destructive	81	0.3%
gdp_growth_metadata	Read	81	0.3%
hdb_carparks_metadata	Read	81	0.3%
singstat_divorces_history	Read	81	0.3%
singstat_household_income_history	Read	81	0.3%
visitor_arrivals_metadata	Read	81	0.3%
dengue_clusters_metadata	Read	80	0.3%
disease_cases_metadata	Read	80	0.3%
sg_acra_formations_by_ssic	Read	80	0.3%
singapore_market_context	Read	80	0.3%
singstat_deaths_history	Read	80	0.3%
singstat_marriages_history	Read	80	0.3%
singstat_wages_history	Read	80	0.3%
analytics_summarize_series	Read	79	0.3%
lta_list_bus_stops	Read	79	0.3%
median_income_metadata	Read	79	0.3%
rail_get_station	Read	79	0.3%
sg_ura_new_sale_pipeline	Execute	78	0.3%
electricity_metadata	Read	78	0.3%
iras_tax_metadata	Read	78	0.3%
population_metadata	Read	78	0.3%
retail_sales_metadata	Read	78	0.3%
sg_coe_latest	Read	78	0.3%
sg_crime_compare	Read	78	0.3%
singapore_ask	Read	78	0.3%
cpi_metadata	Read	77	0.3%
datagov_list_datasets	Read	77	0.3%
employment_sector_metadata	Read	77	0.3%
rail_get_station_exits	Read	77	0.3%
ura_new_sale_pipeline	Execute	76	0.3%
coe_latest	Read	76	0.3%
crime_compare	Read	76	0.3%
crime_metadata	Read	76	0.3%
live_births_metadata	Read	76	0.3%
mas_fx_metadata	Read	76	0.3%
unemployment_metadata	Read	76	0.3%
analytics_visualize_series	Write	76	0.3%
bank_interest_rates_profile	Read	75	0.3%
tourism_receipts_metadata	Read	75	0.3%
sg_iras_tax_mix	Read	74	0.3%
sg_tourism_latest	Read	74	0.3%
acra_get_entity	Read	73	0.3%
ecda_childcare_profile	Read	73	0.3%
hdb_carparks_profile	Read	73	0.3%
hdb_resale_profile	Read	73	0.3%
sg_dataset_query	Read	73	0.3%
singapore_capability_check	Read	73	0.3%
singapore_location_brief	Read	73	0.3%
gdp_growth_profile	Read	72	0.3%
hawker_centres_profile	Read	72	0.3%
iras_tax_mix	Read	72	0.3%
moe_schools_profile	Read	72	0.3%
sg_ask	Read	72	0.3%
ura_private_property_profile	Read	72	0.3%
coe_bidding_profile	Read	71	0.3%
dengue_clusters_profile	Read	71	0.3%
disease_cases_profile	Read	71	0.3%
sg_hdb_carparks_by_type	Write	71	0.3%
iras_tax_profile	Read	70	0.3%
median_income_profile	Read	70	0.3%
retail_sales_profile	Read	70	0.3%
tourism_receipts_profile	Read	70	0.3%
visitor_arrivals_profile	Read	70	0.3%
cpi_profile	Read	69	0.3%
employment_sector_profile	Read	69	0.3%
live_births_profile	Read	69	0.3%
mas_fx_profile	Read	69	0.3%
unemployment_profile	Read	69	0.3%
crime_profile	Read	68	0.3%
electricity_profile	Read	68	0.3%
sg_acra_get_entity	Read	68	0.3%
sg_hdb_carparks_by_address	Read	68	0.3%
acra_profile_entities	Read	67	0.3%
population_profile	Read	67	0.3%
sg_carpark_availability	Read	67	0.3%
sg_moe_school_by_name	Read	67	0.3%
sg_cessations_monthly	Read	66	0.3%
sg_hawker_search	Read	66	0.3%
sg_tourism_history	Read	66	0.3%
sg_acra_search_entities	Read	65	0.2%
sg_formations_history	Read	65	0.2%
sg_formations_latest	Read	65	0.2%
sg_hdb_carpark_lookup	Read	65	0.2%
sg_moe_search_schools	Read	65	0.2%
singstat_business_cessations_monthly_latest	Read	65	0.2%
sg_labour_force	Destructive	64	0.2%
finance_singapore_data_availability	Read	64	0.2%
sg_cpi_history	Read	64	0.2%
sg_ecda_search_centres	Read	64	0.2%
sg_formations_monthly	Read	64	0.2%
sg_hdb_resale_search	Read	64	0.2%
sg_ura_private_txn_history	Read	64	0.2%
sg_ura_private_txn_latest	Read	64	0.2%
sg_visitors_latest	Read	64	0.2%
singstat_business_formations_annual_latest	Read	64	0.2%
singstat_business_formations_monthly_latest	Read	64	0.2%
singstat_merchandise_trade_latest	Read	63	0.2%
singstat_labour_force_latest	Destructive	62	0.2%
disease_cases_list	Read	62	0.2%
sg_disease_list	Read	62	0.2%
sg_hawker_stats	Read	62	0.2%
sg_household_income	Read	62	0.2%
sg_list_datasets	Read	62	0.2%
sg_retail_sales_latest	Read	62	0.2%
singstat_divorces_latest	Read	62	0.2%
singstat_household_income_latest	Read	62	0.2%
rail_network_summary	Read	61	0.2%
sg_births_latest	Read	61	0.2%
sg_dataset_schema	Read	61	0.2%
sg_divorces	Read	61	0.2%
sg_employment_by_sector	Read	61	0.2%
sg_gdp_latest	Read	61	0.2%
sg_median_income_lookup	Read	61	0.2%
sg_trade	Read	61	0.2%
sg_weather_forecast	Read	61	0.2%
singstat_deaths_latest	Read	61	0.2%
singstat_marriages_latest	Read	61	0.2%
singstat_wages_latest	Read	61	0.2%
sg_cpi_latest	Read	60	0.2%
sg_deaths	Read	60	0.2%
sg_electricity_latest	Read	60	0.2%
sg_employment_sector_history	Read	60	0.2%
sg_gdp_history	Read	60	0.2%
sg_iras_collection	Read	60	0.2%
sg_iras_collection_history	Read	60	0.2%
sg_median_income_history	Read	60	0.2%
sg_rainfall	Read	60	0.2%
sg_retail_sales_history	Read	60	0.2%
sg_wages	Read	60	0.2%
sg_births_history	Read	59	0.2%
sg_crime_latest	Read	59	0.2%
sg_dengue_clusters	Read	59	0.2%
sg_marriages	Read	59	0.2%
sg_visitors_history	Read	59	0.2%
singapore_catalog_list	Read	59	0.2%
lta_carpark_availability	Read	58	0.2%
lta_traffic_incidents	Read	58	0.2%
sg_electricity_history	Read	58	0.2%
sg_unemployment_latest	Read	58	0.2%
acra_list_shards	Read	57	0.2%
rail_source_metadata	Read	57	0.2%
sg_air_temperature	Read	57	0.2%
sg_crime_history	Read	57	0.2%
sg_psi	Read	57	0.2%
sg_unemployment_history	Read	57	0.2%
moe_schools_freshness	Read	56	0.2%
sg_population_history	Read	56	0.2%
sg_population_latest	Read	56	0.2%
dengue_clusters_freshness	Read	55	0.2%
ecda_childcare_freshness	Read	55	0.2%
hawker_centres_freshness	Read	55	0.2%
hdb_carparks_freshness	Read	55	0.2%
hdb_resale_freshness	Read	55	0.2%
lta_traffic_images	Read	55	0.2%
tourism_receipts_freshness	Read	55	0.2%
bank_interest_rates_freshness	Read	54	0.2%
coe_bidding_freshness	Read	54	0.2%
disease_cases_freshness	Read	54	0.2%
gdp_growth_freshness	Read	54	0.2%
live_births_freshness	Read	54	0.2%
retail_sales_freshness	Read	54	0.2%
ura_private_property_freshness	Read	54	0.2%
visitor_arrivals_freshness	Read	54	0.2%
cpi_freshness	Read	53	0.2%
electricity_freshness	Read	53	0.2%
employment_sector_freshness	Read	53	0.2%
iras_tax_freshness	Read	53	0.2%
mas_fx_freshness	Read	53	0.2%
median_income_freshness	Read	53	0.2%
unemployment_freshness	Read	53	0.2%
crime_freshness	Read	52	0.2%
lta_taxi_availability	Read	52	0.2%
population_freshness	Read	52	0.2%
singapore_prompt_categories	Read	51	0.2%
singapore_cache_info	Read	50	0.2%

// CUT THE COST

Your agent uses a handful of these tools. It pays for all 291.

You don't need all 291 of those definitions in the window. PolicyLayer is an MCP gateway that sits in front of Mcp Singapore: 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 291 tools (no gateway)	26,187 tokens	—
3 granted tools	~270 tokens	−99%
5 granted tools	~450 tokens	−98%
10 granted tools	~900 tokens	−97%

The risk dividend: 3 of these 291 tools are critical-risk (destructive or financial) and cost 207 tokens (1% 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 Mcp Singapore — 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 MCP SINGAPORE TOKEN COST →

Instant setup, no code required.

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

//FAQ

Mcp Singapore token-cost questions.

How many tokens does the Mcp Singapore MCP server use?+

Its 291 tool definitions total 26,187 tokens — 13% 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 Mcp Singapore 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 Mcp Singapore's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Mcp Singapore 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 270 tokens, a 99% 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 21-10-2026 from the PolicyLayer scan database over all 291 catalogued Mcp Singapore tools. Counts refresh with every site build.