// MCP TOKEN COST

The SendIt MCP server costs 23,650 tokens before the first call.

Connect SendIt and its 148 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 SendIt MCP server's tool definitions consume 23,650 tokens — 12× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 148 tools · 23,650 tokens · 12% of 200k · 2.4% 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 2.4%

Corpus context: SendIt ranks #42 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 23,650 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 160 tokens per tool.

Tool	Category	Tokens	% of server
schedule_content	Write	1,261	5.3%
publish_content	Write	939	4.0%
schedule_content_advanced	Write	874	3.7%
validate_content	Read	823	3.5%
generate_content	Write	420	1.8%
edit_scheduled_post	Write	418	1.8%
create_workflow	Write	389	1.6%
publish_ai	Write	367	1.6%
create_brand_voice	Write	366	1.5%
retry_scheduled_post	Write	353	1.5%
bulk_update_calendar_events	Write	335	1.4%
get_scheduled_posts	Read	316	1.3%
schedule_multilang_content	Write	308	1.3%
save_to_library	Write	307	1.3%
preview_content	Write	300	1.3%
list_calendar_events	Read	297	1.3%
bulk_schedule	Write	295	1.2%
reply_to_conversation	Write	291	1.2%
update_agent_policy	Write	290	1.2%
score_content	Execute	288	1.2%
suggest_next_schedule_time	Write	286	1.2%
create_library_item	Write	280	1.2%
get_analytics	Read	264	1.1%
create_ad_campaign	Write	261	1.1%
edit_post	Write	235	1.0%
create_ad_creative	Write	234	1.0%
repurpose_url	Read	233	1.0%
update_calendar_event	Write	229	1.0%
upload_media	Write	228	1.0%
get_ad_performance	Read	216	0.9%
list_media_assets	Read	214	0.9%
suggest_queue_slots	Write	211	0.9%
get_calendar_gaps	Read	204	0.9%
autopilot_set_goal	Write	203	0.9%
get_unified_analytics	Read	202	0.9%
update_library_item	Write	199	0.8%
get_attribution_report	Read	191	0.8%
request_project_content	Write	189	0.8%
get_audit_log	Read	187	0.8%
generate_ai_media	Write	184	0.8%
delete_post	Destructive	182	0.8%
create_scheduled_report	Write	179	0.8%
list_conversations	Read	177	0.7%
list_inbox	Read	176	0.7%
list_agents	Read	175	0.7%
publish_from_library	Write	173	0.7%
invoke_agent	Execute	171	0.7%
get_calendar_recommendations	Read	170	0.7%
get_best_times	Read	164	0.7%
get_platform_requirements	Read	164	0.7%
create_upload_session	Write	160	0.7%
create_project	Write	156	0.7%
generate_post_bundle	Write	156	0.7%
duplicate_scheduled_post	Write	150	0.6%
save_project_content_to_library	Write	149	0.6%
connect_connector	Write	146	0.6%
update_recurring_series	Write	144	0.6%
list_agent_runs	Read	142	0.6%
update_conversation	Write	142	0.6%
execute_connector_operation	Execute	141	0.6%
list_library	Read	141	0.6%
escalate_to_support	Read	140	0.6%
autopilot_approve_plan	Write	135	0.6%
get_upload_session	Read	134	0.6%
list_connectors	Read	132	0.6%
get_platform_settings_schema	Read	131	0.6%
get_unified_ad_report	Read	130	0.5%
create_goal	Write	130	0.5%
list_dead_letter	Read	126	0.5%
connect_platform	Write	126	0.5%
autopilot_generate_plan	Write	121	0.5%
get_post_analytics	Read	119	0.5%
get_benchmark_comparison	Read	118	0.5%
update_ad_campaign	Write	116	0.5%
get_producthunt_analytics	Read	114	0.5%
get_scheduled_post	Read	114	0.5%
critique_post	Write	111	0.5%
summarize_mentions	Write	110	0.5%
approve_post	Write	108	0.5%
import_project_source	Write	107	0.5%
list_projects	Read	105	0.4%
draft_reply	Write	105	0.4%
reject_post	Write	105	0.4%
get_agent_policies	Read	104	0.4%
get_anomaly_alerts	Read	104	0.4%
list_recurring_series	Read	104	0.4%
complete_goal	Write	101	0.4%
requeue_dead_letter	Write	101	0.4%
run_analytics_report	Execute	99	0.4%
reply_to_comment	Write	99	0.4%
cancel_recurring_series	Destructive	97	0.4%
get_approval_comments	Read	96	0.4%
upvote_product	Write	91	0.4%
trigger_scheduled_post	Execute	87	0.4%
list_workflow_runs	Read	82	0.3%
delete_scheduled_post	Destructive	81	0.3%
get_library_item	Read	81	0.3%
delete_library_item	Destructive	80	0.3%
list_ad_campaigns	Read	80	0.3%
list_workflows	Read	80	0.3%
get_connector_capabilities	Read	78	0.3%
list_brand_voices	Read	77	0.3%
update_workflow	Write	77	0.3%
analyze_project	Read	76	0.3%
get_tiktok_creator_info	Read	74	0.3%
list_ad_accounts	Read	74	0.3%
generate_project_strategy	Write	74	0.3%
check_ai_media_status	Read	73	0.3%
get_project_analysis	Read	72	0.3%
list_connected_accounts	Read	72	0.3%
trigger_workflow	Execute	69	0.3%
get_inbox_summary	Read	69	0.3%
list_connected_connectors	Read	69	0.3%
autopilot_execute_plan	Execute	67	0.3%
list_analytics_reports	Read	66	0.3%
set_default_brand_voice	Write	66	0.3%
disconnect_connector	Write	65	0.3%
autopilot_get_progress	Read	62	0.3%
list_teams	Read	57	0.2%
get_workflow_run	Read	51	0.2%
connect_whop	Write	50	0.2%
get_agent_run	Read	49	0.2%
get_connector_health	Read	48	0.2%
get_conversation	Read	45	0.2%
connect_nostr	Write	44	0.2%
delete_workflow	Destructive	42	0.2%
list_pending_approvals	Read	40	0.2%
connect_bluesky	Write	38	0.2%
connect_lemmy	Write	38	0.2%
connect_devto	Write	37	0.2%
connect_hashnode	Write	37	0.2%
connect_telegram	Write	36	0.2%
autopilot_list_goals	Read	34	0.1%
connect_dribbble	Write	34	0.1%
connect_linkedin-page	Write	34	0.1%
connect_gmb	Write	33	0.1%
connect_mastodon	Write	33	0.1%
connect_linkedin	Write	32	0.1%
connect_producthunt	Write	32	0.1%
connect_tiktok	Write	32	0.1%
connect_youtube	Write	32	0.1%
connect_facebook	Write	31	0.1%
connect_instagram	Write	31	0.1%
connect_pinterest	Write	31	0.1%
connect_discord	Write	30	0.1%
connect_slack	Write	30	0.1%
connect_threads	Write	30	0.1%
connect_x	Write	30	0.1%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 148.

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 (160 tokens each).

Grant scope	Definition cost	Reduction
All 148 tools (no gateway)	23,650 tokens	—
3 granted tools	~479 tokens	−98%
5 granted tools	~799 tokens	−97%
10 granted tools	~1,598 tokens	−93%

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

//FAQ

SendIt token-cost questions.

How many tokens does the SendIt MCP server use?+

Its 148 tool definitions total 23,650 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 SendIt 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 SendIt's token usage?+

Expose fewer tools. A PolicyLayer grant scopes SendIt 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 479 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 148 catalogued SendIt tools. Counts refresh with every site build.