// MCP TOKEN COST

The Rocket Plus MCP server costs 7,408 tokens before the first call.

Connect Rocket Plus and its 106 tool definitions are loaded into the model's context on every request — 3.7% of a 200k window spent before your agent does anything.

QUICK ANSWER The Rocket Plus MCP server's tool definitions consume 7,408 tokens — 3.9× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 106 tools · 7,408 tokens · 3.7% of 200k · 0.7% 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 3.7%

1M WINDOW 0.7%

Corpus context: Rocket Plus ranks #795 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 7,408 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 70 tokens per tool.

Tool	Category	Tokens	% of server
recurring_tasks_create	Write	201	2.7%
course_generate	Write	154	2.1%
content_generate	Write	142	1.9%
apex_create_test	Write	130	1.8%
pdl_search_people	Read	129	1.7%
google_calendar_create_event	Write	126	1.7%
recurring_tasks_update	Write	119	1.6%
pdl_search_companies	Read	113	1.5%
pdl_enrich_person	Read	110	1.5%
create_contact	Write	109	1.5%
canvas_create	Write	106	1.4%
create_opportunity	Write	103	1.4%
dashboard_add_widget	Write	102	1.4%
insights_analyze	Read	100	1.3%
game_create	Write	100	1.3%
agent_execute	Execute	99	1.3%
google_gmail_send	Write	98	1.3%
tasks_analyze	Read	95	1.3%
update_contact	Write	91	1.2%
gcloud_functions_invoke	Execute	90	1.2%
google_sheets_append	Read	90	1.2%
pdl_auto_enrich_contact	Read	90	1.2%
rss_add_feed	Write	90	1.2%
rocketflow_deploy	Execute	89	1.2%
form_create	Write	89	1.2%
rss_generate_post	Write	86	1.2%
gcloud_storage_upload	Write	85	1.1%
tasks_complete	Write	85	1.1%
google_drive_upload	Write	84	1.1%
canvas_generate	Write	82	1.1%
google_sheets_write	Write	80	1.1%
gcloud_bigquery_query	Read	79	1.1%
mcp_call_server	Read	79	1.1%
content_rewrite	Read	78	1.1%
get_contacts	Read	76	1.0%
focus_create_session	Write	76	1.0%
seo_generate_meta	Write	76	1.0%
insights_predict	Read	75	1.0%
send_email	Write	75	1.0%
webhook_create	Write	75	1.0%
content_ideas	Read	74	1.0%
google_calendar_events	Read	74	1.0%
tasks_sync_google	Read	74	1.0%
tasks_sync_sheets	Read	74	1.0%
google_sheets_read	Read	72	1.0%
visitors_history	Read	72	1.0%
dashboard_create	Write	71	1.0%
form_get_submissions	Read	70	0.9%
pdl_enrich_company	Read	68	0.9%
research_competitors	Read	68	0.9%
research_trends	Read	68	0.9%
insights_report	Read	67	0.9%
gcloud_apis_enable	Write	67	0.9%
tasks_sync_calendar	Write	67	0.9%
google_gmail_search	Read	66	0.9%
focus_get_tasks	Read	65	0.9%
apex_optimize	Read	63	0.9%
recurring_tasks_history	Read	63	0.9%
seo_keyword_research	Read	63	0.9%
api_test	Read	62	0.8%
cro_heatmap	Read	62	0.8%
research_ideas	Read	61	0.8%
rss_get_items	Read	61	0.8%
seo_analyze	Read	61	0.8%
canvas_export	Write	60	0.8%
add_tags	Write	59	0.8%
google_sheets_create	Write	59	0.8%
send_sms	Write	58	0.8%
cro_analyze	Read	57	0.8%
google_drive_create_folder	Write	57	0.8%
course_outline	Read	56	0.8%
get_opportunities	Read	55	0.7%
gcloud_functions_list	Read	54	0.7%
tasks_sync_ghl	Read	53	0.7%
google_drive_search	Read	52	0.7%
recurring_tasks_run_now	Execute	51	0.7%
google_calendar_freebusy	Read	51	0.7%
google_drive_list	Read	51	0.7%
recurring_tasks_list	Read	50	0.7%
agent_status	Read	48	0.6%
rocketflow_validate	Read	48	0.6%
apex_get_results	Read	46	0.6%
google_tasks_get	Read	46	0.6%
mcp_list_servers	Read	46	0.6%
tasks_get_lists	Read	46	0.6%
recurring_tasks_delete	Destructive	45	0.6%
api_list_endpoints	Read	45	0.6%
rocketflow_templates	Read	45	0.6%
gcloud_storage_list	Read	44	0.6%
visitors_active	Read	42	0.6%
game_stats	Read	41	0.6%
gcloud_apis_list	Read	40	0.5%
gcloud_bigquery_datasets	Read	39	0.5%
webhook_list	Read	39	0.5%
gcloud_iam_service_accounts	Read	38	0.5%
gcloud_storage_buckets	Read	38	0.5%
get_location_stats	Read	34	0.5%
list_mods	Read	31	0.4%
get_usage	Read	28	0.4%
gcloud_projects_list	Read	27	0.4%
gcloud_status	Read	27	0.4%
google_calendar_list	Read	27	0.4%
google_tasks_status	Read	27	0.4%
google_workspace_status	Read	27	0.4%
agent_list	Read	26	0.4%
google_gmail_labels	Read	26	0.4%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 106.

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 (70 tokens each).

Grant scope	Definition cost	Reduction
All 106 tools (no gateway)	7,408 tokens	—
3 granted tools	~210 tokens	−97%
5 granted tools	~349 tokens	−95%
10 granted tools	~699 tokens	−91%

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

//FAQ

Rocket Plus token-cost questions.

How many tokens does the Rocket Plus MCP server use?+

Its 106 tool definitions total 7,408 tokens — 3.7% 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 Rocket Plus 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 Rocket Plus's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Rocket Plus 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 210 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.

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 106 catalogued Rocket Plus tools. Counts refresh with every site build.