// MCP TOKEN COST

The Wavix MCP server costs 24,917 tokens before the first call.

Connect Wavix and its 122 tool definitions are loaded into the model's context on every request — 12% of a 200k window spent before your agent does anything.

QUICK ANSWER The Wavix MCP server's tool definitions consume 24,917 tokens — 13× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 122 tools · 24,917 tokens · 12% of 200k · 2.5% 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 12%

1M WINDOW 2.5%

Corpus context: Wavix ranks #38 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 24,917 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 204 tokens per tool.

Tool	Category	Tokens	% of server
api_keys_create	Write	1,611	6.5%
ten_dlc_brand_vettings_import	Write	1,117	4.5%
ten_dlc_brand_campaigns_update	Write	1,056	4.2%
cdrs_search	Read	1,032	4.1%
sip_trunks_update	Write	940	3.8%
ten_dlc_brand_campaigns_create	Write	931	3.7%
sip_trunks_create	Write	913	3.7%
billing_transactions_list	Read	768	3.1%
ten_dlc_brands_create	Write	757	3.0%
ten_dlc_brands_update	Write	740	3.0%
sms_and_mms_messages_list	Read	397	1.6%
cdrs_list_all	Read	384	1.5%
cdrs_list	Read	381	1.5%
sms_and_mms_messages_list_all	Read	379	1.5%
profile_update	Write	362	1.5%
ten_dlc_brands_list	Read	350	1.4%
call_control_collect	Read	339	1.4%
ten_dlc_brand_appeals_create	Write	322	1.3%
my_numbers_destinations_update	Write	317	1.3%
sms_and_mms_messages_send	Write	312	1.3%
call_control_answer	Read	293	1.2%
ten_dlc_brand_campaigns_list	Read	286	1.1%
two_fa_verification_create	Write	285	1.1%
call_recording_list	Read	280	1.1%
ten_dlc_campaigns_list	Read	275	1.1%
ten_dlc_brand_vetting_appeals_create	Write	274	1.1%
sub_accounts_transactions_list	Read	270	1.1%
sms_and_mms_sender_ids_create	Write	263	1.1%
ten_dlc_campaigns_nudge	Read	262	1.1%
speech_analytics_create	Write	260	1.0%
call_control_streams_create	Write	242	1.0%
voice_campaigns_create	Write	230	0.9%
ten_dlc_brand_usecase_qualify	Read	220	0.9%
call_control_create	Write	214	0.9%
my_numbers_list	Read	199	0.8%
my_numbers_papers_upload	Write	193	0.8%
number_validator_create_bulk	Write	189	0.8%
sub_accounts_update	Write	188	0.8%
link_shortener_metrics_list	Read	187	0.8%
sms_and_mms_opt_outs_list	Read	185	0.7%
my_numbers_update	Write	184	0.7%
link_shortener_create	Write	173	0.7%
call_webhooks_create	Write	161	0.6%
sms_and_mms_opt_outs_create	Write	149	0.6%
two_fa_sessions_list	Read	146	0.6%
sub_accounts_create	Write	137	0.5%
ten_dlc_brand_vettings_create	Write	137	0.5%
buy_numbers_list	Read	134	0.5%
cdrs_retranscribe	Read	125	0.5%
ten_dlc_campaign_numbers_link	Read	125	0.5%
two_fa_verification_resend	Read	125	0.5%
speech_analytics_retranscribe	Read	124	0.5%
webrtc_tokens_create	Write	124	0.5%
buy_region_cities_list	Read	119	0.5%
ten_dlc_campaign_numbers_unlink	Read	118	0.5%
ten_dlc_subscriptions_create	Write	112	0.4%
my_numbers_delete	Destructive	105	0.4%
two_fa_verification_cancel	Destructive	104	0.4%
ten_dlc_brand_evidence_get	Read	99	0.4%
ten_dlc_brand_evidence_upload	Write	99	0.4%
ten_dlc_brand_evidence_delete	Destructive	98	0.4%
number_validator_get	Read	97	0.4%
sip_trunks_list	Read	97	0.4%
two_fa_verification_check	Read	96	0.4%
ten_dlc_brand_campaigns_delete	Destructive	95	0.4%
buy_regions_list	Read	95	0.4%
speech_analytics_file_get	Read	93	0.4%
buy_cities_list	Read	92	0.4%
ten_dlc_campaign_numbers_list	Read	88	0.4%
cdrs_get	Read	86	0.3%
call_recording_get	Read	85	0.3%
sub_accounts_list	Read	83	0.3%
call_webhooks_delete	Destructive	82	0.3%
billing_invoices_download	Read	81	0.3%
billing_invoices_list	Read	80	0.3%
ten_dlc_brand_campaigns_get	Read	80	0.3%
cart_checkout	Write	79	0.3%
my_numbers_sms_update	Write	79	0.3%
webrtc_tokens_update	Write	79	0.3%
call_control_audio_play	Read	78	0.3%
two_fa_events_list	Read	78	0.3%
sub_accounts_get	Read	77	0.3%
call_control_update	Write	75	0.3%
webrtc_tokens_delete	Destructive	74	0.3%
number_validator_results_get	Read	71	0.3%
sms_and_mms_messages_get	Read	70	0.3%
speech_analytics_get	Read	70	0.3%
sms_and_mms_sender_ids_delete	Destructive	69	0.3%
ten_dlc_brand_vetting_appeals_list	Read	69	0.3%
cart_add	Write	69	0.3%
call_control_streams_delete	Destructive	67	0.3%
cart_remove	Destructive	67	0.3%
webrtc_tokens_list	Read	66	0.3%
cdrs_transcription_get	Read	65	0.3%
ten_dlc_brand_vettings_list	Read	65	0.3%
ten_dlc_brands_delete	Destructive	64	0.3%
ten_dlc_brand_evidence_list	Read	63	0.3%
ten_dlc_brand_appeals_list	Read	59	0.2%
api_keys_deactivate	Read	58	0.2%
buy_countries_list	Read	58	0.2%
ten_dlc_subscriptions_delete	Destructive	57	0.2%
ten_dlc_brands_get	Read	56	0.2%
api_keys_activate	Write	56	0.2%
sip_trunks_get	Read	55	0.2%
voice_campaigns_get	Read	55	0.2%
sip_trunks_delete	Destructive	54	0.2%
webrtc_tokens_get	Read	53	0.2%
call_webhooks_list	Read	52	0.2%
call_recording_delete	Destructive	51	0.2%
my_numbers_get	Read	51	0.2%
sms_and_mms_sender_ids_get	Read	50	0.2%
call_control_audio_stop	Execute	49	0.2%
api_keys_list	Read	48	0.2%
api_keys_delete	Destructive	47	0.2%
call_control_delete	Destructive	45	0.2%
call_control_get	Read	44	0.2%
ten_dlc_subscriptions_list	Read	38	0.2%
sms_and_mms_sender_ids_list	Read	35	0.1%
profile_config_get	Read	32	0.1%
call_control_list	Read	31	0.1%
profile_get	Read	31	0.1%
cart_get	Read	27	0.1%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 122.

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 (204 tokens each).

Grant scope	Definition cost	Reduction
All 122 tools (no gateway)	24,917 tokens	—
3 granted tools	~613 tokens	−98%
5 granted tools	~1,021 tokens	−96%
10 granted tools	~2,042 tokens	−92%

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

//FAQ

Wavix token-cost questions.

How many tokens does the Wavix MCP server use?+

Its 122 tool definitions total 24,917 tokens — 12% 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 Wavix 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 Wavix's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Wavix 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 613 tokens, a 98% 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 122 catalogued Wavix tools. Counts refresh with every site build.