// MCP TOKEN COST

The Vee3 MCP server costs 59,899 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 Vee3 MCP server's 249 tool definitions consume 59,899 tokens — 30% of a 200k context window, and 31× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS tiktoken o200k_base · rank #27 of 8,377 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 30%

1M WINDOW 6.0%

Corpus context: Vee3 ranks #27 of 8,377 measured MCP servers by definition cost. The median is 1,905 tokens, p90 is 12,638, and the heaviest (Ainumbers Mcp Apps) is 327,608 — 164% 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 59,899 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 241 tokens per tool.

Tool	Category	Tokens	% of server
x-twitter.create_post	Write	1,067	1.8%
x-twitter.reply_to_post	Write	937	1.6%
youtube.search	Read	697	1.2%
website-screenshots.capture	Execute	688	1.1%
google-maps.nearby_search	Read	666	1.1%
google-maps.search	Read	606	1.0%
clerk.update_enterprise_connection	Write	587	1.0%
clerk.create_enterprise_connection	Write	571	1.0%
x-twitter.edit_post	Write	510	0.9%
google-trends.interest_by_region	Read	505	0.8%
files.upload_file	Write	493	0.8%
google-trends.related_queries	Read	484	0.8%
clerk.update_instance_settings	Write	471	0.8%
clerk.update_jwt_template	Write	469	0.8%
google-trends.interest_over_time	Read	467	0.8%
clerk.update_instance_organization_settings	Write	466	0.8%
clerk.create_api_key	Write	460	0.8%
google-maps.place_reviews	Read	432	0.7%
clerk.create_jwt_template	Write	420	0.7%
clerk.list_organization_domains	Read	401	0.7%
google-search.keyword_traffic_insights	Read	393	0.7%
clerk.update_oauth_application	Write	388	0.6%
clerk.create_m2m_token	Write	385	0.6%
clerk.update_api_key	Write	376	0.6%
clerk.list_api_keys	Read	373	0.6%
clerk.create_oauth_application	Write	370	0.6%
clerk.list_sessions	Read	365	0.6%
clerk.update_instance_restrictions	Write	361	0.6%
google-maps.place_photos	Read	360	0.6%
clerk.update_organization	Write	359	0.6%
domains.list_tlds	Read	358	0.6%
clerk.create_actor_token	Execute	358	0.6%
clerk.list_m2m_tokens	Read	356	0.6%
clerk.create_admin_portal_link_token	Write	352	0.6%
clerk.create_organization	Write	349	0.6%
google-search.url_traffic_insights	Read	348	0.6%
domains.register	Financial	341	0.6%
clerk.list_waitlist_entries	Read	338	0.6%
rotten-tomatoes.tv_show_season_reviews	Read	334	0.6%
clerk.list_enterprise_connection_test_runs	Read	328	0.5%
x-twitter.get_bookmarks	Read	327	0.5%
clerk.create_phone_number	Write	326	0.5%
clerk.update_organization_role	Write	318	0.5%
domains.add_dns_host	Write	318	0.5%
clerk.update_user_metadata	Read	317	0.5%
clerk.list_enterprise_connections	Read	313	0.5%
clerk.list_machines	Read	309	0.5%
google-maps.place_details	Read	307	0.5%
tiktok.search_videos	Read	306	0.5%
clerk.list_organization_invitations	Read	301	0.5%
clerk.create_organization_role	Write	299	0.5%
clerk.update_phone_number	Write	295	0.5%
clerk.update_user	Write	295	0.5%
clerk.create_machine	Write	294	0.5%
clerk.list_organization_roles	Read	293	0.5%
clerk.list_organization_permissions	Read	290	0.5%
clerk.update_organization_membership_metadata	Read	290	0.5%
tiktok.user_videos	Read	287	0.5%
clerk.create_organization_domain	Write	284	0.5%
rotten-tomatoes.movie_reviews	Read	283	0.5%
tiktok.download_video	Write	280	0.5%
clerk.create_email_address	Write	276	0.5%
clerk.list_organization_memberships	Read	275	0.5%
clerk.list_user_organization_memberships	Read	272	0.5%
youtube.video_comments	Read	271	0.5%
clerk.create_sign_in_token	Write	271	0.5%
clerk.upload_oauth_application_logo	Write	270	0.5%
meta-tools.request_feature	Write	267	0.4%
x-twitter.user_timeline	Write	267	0.4%
clerk.create_session_token	Execute	265	0.4%
clerk.update_organization_metadata	Read	263	0.4%
clerk.create_user	Write	261	0.4%
clerk.update_organization_domain	Write	261	0.4%
clerk.list_organizations	Read	257	0.4%
clerk.list_invitations	Read	256	0.4%
clerk.revoke_admin_portal_link_token	Destructive	254	0.4%
meta-tools.report_bug	Write	253	0.4%
clerk.update_machine	Write	253	0.4%
clerk.update_email_address	Write	252	0.4%
clerk.rotate_machine_secret_key	Write	248	0.4%
clerk.create_organization_membership	Write	248	0.4%
x-twitter.repost_post	Write	248	0.4%
clerk.create_organization_invitations	Read	247	0.4%
meta-tools.search	Read	247	0.4%
clerk.replace_user_metadata	Write	246	0.4%
clerk.update_domain	Write	241	0.4%
files.download_file	Read	239	0.4%
clerk.update_organization_membership	Write	239	0.4%
clerk.create_allowlist_identifier	Write	239	0.4%
x-twitter.unrepost_post	Write	238	0.4%
tiktok.download_music_from_video	Write	237	0.4%
x-twitter.delete_bookmark	Destructive	236	0.4%
clerk.change_instance_domain	Read	236	0.4%
tiktok.user_reposts	Read	234	0.4%
x-twitter.create_bookmark	Write	234	0.4%
clerk.revoke_api_key	Destructive	233	0.4%
tiktok.download_music	Write	233	0.4%
clerk.add_domain	Write	233	0.4%
clerk.update_organization_permission	Write	233	0.4%
x-twitter.delete_post	Destructive	233	0.4%
clerk.get_development_api_keys	Read	232	0.4%
tiktok.video_comments	Read	229	0.4%
clerk.update_instance_oauth_settings	Write	229	0.4%
youtube.channel_search	Read	226	0.4%
clerk.create_waitlist_entries	Write	225	0.4%
clerk.invite_waitlist_entry	Write	222	0.4%
clerk.rotate_oauth_application_secret	Write	221	0.4%
youtube.channel_videos	Read	219	0.4%
clerk.revoke_organization_invitation	Destructive	217	0.4%
clerk.list_users	Read	216	0.4%
tiktok.search_photos	Read	213	0.4%
clerk.delete_user_external_account	Destructive	212	0.4%
clerk.create_organization_permission	Write	212	0.4%
clerk.list_redirect_urls	Read	210	0.4%
clerk.revoke_m2m_token	Destructive	209	0.3%
tiktok.comment_replies	Read	209	0.3%
clerk.create_machine_scope	Write	209	0.3%
clerk.delete_machine_scope	Destructive	208	0.3%
clerk.delete_organization_membership	Destructive	208	0.3%
clerk.delete_user_web3_wallet	Destructive	208	0.3%
clerk.get_api_key_secret	Read	208	0.3%
x-twitter.user_info	Write	208	0.3%
clerk.get_organization_invitation	Read	207	0.3%
rotten-tomatoes.tv_show_episode	Read	207	0.3%
clerk.create_invitations	Write	206	0.3%
tiktok.search_users	Read	206	0.3%
clerk.delete_user_passkey	Destructive	205	0.3%
clerk.revoke_sign_in_token	Destructive	205	0.3%
x-twitter.search	Read	205	0.3%
clerk.get_jwt_template	Read	204	0.3%
clerk.revoke_actor_token	Destructive	200	0.3%
clerk.list_allowlist_identifiers	Read	200	0.3%
clerk.list_blocklist_identifiers	Read	200	0.3%
clerk.create_session	Write	200	0.3%
clerk.create_enterprise_connection_test_run	Execute	199	0.3%
clerk.get_machine_secret_key	Read	199	0.3%
domains.renew	Financial	198	0.3%
clerk.create_blocklist_identifier	Write	197	0.3%
files.extend_file	Write	197	0.3%
clerk.verify_organization_domain_ownership	Write	196	0.3%
clerk.get_session	Read	195	0.3%
clerk.update_instance_communication_settings	Write	195	0.3%
domains.whois	Read	195	0.3%
clerk.get_enterprise_connection	Read	194	0.3%
clerk.unban_users	Write	193	0.3%
clerk.delete_jwt_template	Destructive	192	0.3%
clerk.get_sign_up	Read	191	0.3%
domains.dns_records	Read	191	0.3%
files.list_uploaded_files	Write	191	0.3%
clerk.delete_api_key	Destructive	190	0.3%
clerk.count_users	Read	190	0.3%
youtube.video_details	Read	189	0.3%
clerk.delete_oauth_application	Destructive	188	0.3%
clerk.ban_users	Read	188	0.3%
clerk.delete_enterprise_connection	Destructive	187	0.3%
clerk.get_api_key	Read	187	0.3%
clerk.delete_phone_number	Destructive	186	0.3%
clerk.delete_waitlist_entry	Destructive	186	0.3%
youtube.channel_details	Read	186	0.3%
clerk.get_oauth_application	Read	185	0.3%
clerk.reject_waitlist_entry	Read	185	0.3%
clerk.disable_user_mfa	Write	184	0.3%
clerk.delete_organization	Destructive	183	0.3%
clerk.delete_organization_domain	Destructive	183	0.3%
clerk.delete_email_address	Destructive	182	0.3%
clerk.revoke_session	Destructive	182	0.3%
domains.renewal_quote	Read	182	0.3%
clerk.get_organization	Read	181	0.3%
clerk.get_phone_number	Read	181	0.3%
clerk.delete_domain	Destructive	180	0.3%
tiktok.music_videos	Read	180	0.3%
clerk.delete_redirect_url	Destructive	179	0.3%
clerk.get_organization_role	Read	179	0.3%
tiktok.user_followers	Read	179	0.3%
tiktok.user_following	Read	179	0.3%
clerk.delete_allowlist_identifier	Destructive	178	0.3%
clerk.delete_blocklist_identifier	Destructive	178	0.3%
clerk.revoke_invitation	Destructive	177	0.3%
clerk.get_email_address	Read	177	0.3%
clerk.delete_organization_role	Destructive	175	0.3%
clerk.delete_user_totp	Destructive	175	0.3%
clerk.get_machine	Read	175	0.3%
clerk.get_redirect_url	Read	175	0.3%
clerk.delete_user_backup_codes	Destructive	174	0.3%
files.move_file	Write	174	0.3%
clerk.list_oauth_applications	Read	173	0.3%
clerk.delete_machine	Destructive	172	0.3%
clerk.delete_organization_permission	Destructive	172	0.3%
clerk.get_user	Read	172	0.3%
youtube.search_autocomplete	Read	172	0.3%
clerk.create_svix_dashboard_url	Read	171	0.3%
clerk.delete_user	Destructive	170	0.3%
clerk.list_jwt_templates	Read	170	0.3%
clerk.create_redirect_url	Write	170	0.3%
rotten-tomatoes.tv_show_season	Read	167	0.3%
domains.metrics	Read	166	0.3%
seo.backlinks	Read	165	0.3%
rotten-tomatoes.search	Read	164	0.3%
x-twitter.tweet_replies	Read	164	0.3%
clerk.get_instance	Read	160	0.3%
clerk.create_svix_app	Write	151	0.3%
clerk.get_jwks	Write	151	0.3%
domains.tld_details	Read	150	0.3%
clerk.delete_svix_app	Destructive	147	0.2%
meta-tools.describe	Read	147	0.2%
seo.keyword_metrics	Read	145	0.2%
x-twitter.connected_accounts	Read	144	0.2%
clerk.get_instance_organization_settings	Read	142	0.2%
domains.lookup	Read	142	0.2%
clerk.get_instance_communication_settings	Read	141	0.2%
rotten-tomatoes.tv_show_details	Read	141	0.2%
clerk.get_instance_oauth_settings	Read	140	0.2%
clerk.list_domains	Read	139	0.2%
clerk.get_connected_accounts	Read	138	0.2%
clerk.get_instance_settings	Read	138	0.2%
google-trends.suggestions	Read	137	0.2%
rotten-tomatoes.celebrity_details	Read	137	0.2%
rotten-tomatoes.movie_details	Read	137	0.2%
domains.registration_quote	Read	136	0.2%
tiktok.music_details	Read	136	0.2%
tiktok.user_info	Read	135	0.2%
tiktok.video_details	Read	134	0.2%
seo.url_metrics	Read	132	0.2%
domains.list_managed	Read	131	0.2%
rotten-tomatoes.tv_show_cast_and_crew	Read	130	0.2%
tiktok.for_you_feed	Read	130	0.2%
domains.check_availability	Read	129	0.2%
youtube.playlist_details	Read	129	0.2%
domains.remove_dns_host	Destructive	128	0.2%
rotten-tomatoes.movie_cast_and_crew	Read	127	0.2%
seo.basic_metrics	Read	121	0.2%
google-maps.review_details	Read	120	0.2%
google-trends.regions	Read	119	0.2%
domains.rdap	Read	118	0.2%
files.delete_file	Destructive	117	0.2%
x-twitter.tweet_info	Read	117	0.2%
meta-tools.list_group_tools	Read	115	0.2%
domains.get_dns_hosts	Read	101	0.2%
domains.get_managed	Read	99	0.2%
google-search.languages	Read	94	0.2%
google-trends.categories	Read	94	0.2%
meta-tools.list_all	Read	93	0.2%
google-search.autocomplete	Read	90	0.2%
google-search.locations	Read	88	0.1%
meta-tools.token_balance	Read	88	0.1%
google-maps.languages	Read	84	0.1%
google-maps.place_types	Read	81	0.1%
meta-tools.list_groups	Read	72	0.1%
seo.country_codes	Read	72	0.1%

// CUT THE COST

Your agent uses a handful of these tools. It pays for all 249.

You don't need all 249 of those definitions in the window. PolicyLayer is an MCP gateway that sits in front of Vee3: 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 249 tools (no gateway)	59,899 tokens	—
3 granted tools	~722 tokens	−99%
5 granted tools	~1,203 tokens	−98%
10 granted tools	~2,406 tokens	−96%

The risk dividend: 39 of these 249 tools are critical-risk (destructive or financial) and cost 7,548 tokens (13% 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 Vee3 — 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 VEE3 TOKEN COST →

Instant setup, no code required.

Model your own stack in the token-cost calculator, or see the Vee3 policy for what a sensible grant looks like.

//FAQ

Vee3 token-cost questions.

How many tokens does the Vee3 MCP server use?+

Its 249 tool definitions total 59,899 tokens — 30% 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 Vee3 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 Vee3's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Vee3 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 722 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 06-09-2026 from the PolicyLayer scan database over all 249 catalogued Vee3 tools. Counts refresh with every site build.