// MCP TOKEN COST

The BIGHUB MCP server costs 11,747 tokens before the first call.

Connect BIGHUB and its 125 tool definitions are loaded into the model's context on every request — 5.9% of a 200k window spent before your agent does anything.

QUICK ANSWER The BIGHUB MCP server's tool definitions consume 11,747 tokens — 6.2× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 125 tools · 11,747 tokens · 5.9% of 200k · 1.2% 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 5.9%

1M WINDOW 1.2%

Corpus context: BIGHUB ranks #138 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 11,747 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 94 tokens per tool.

Tool	Category	Tokens	% of server
bighub_ingest_event	Read	593	5.0%
bighub_outcomes_report	Read	299	2.5%
bighub_cases_create	Write	277	2.4%
bighub_actions_memory_recommendations	Read	251	2.1%
bighub_features_compute_batch	Read	239	2.0%
bighub_constraints_apply_patch	Write	206	1.8%
bighub_actions_evaluate_batch	Read	201	1.7%
bighub_retrieval_compare	Read	195	1.7%
bighub_cases_precedents	Read	194	1.7%
bighub_http_request	Read	187	1.6%
bighub_cases_list	Read	166	1.4%
bighub_precedents_query	Read	162	1.4%
bighub_retrieval_index_case	Read	157	1.3%
bighub_actions_evaluate	Read	155	1.3%
bighub_cases_report_outcome	Read	154	1.3%
bighub_actions_memory_ingest	Read	149	1.3%
bighub_ingest_reconcile	Read	149	1.3%
bighub_retrieval_query	Read	140	1.2%
bighub_features_compute	Read	137	1.2%
bighub_retrieval_query_explained	Read	136	1.2%
bighub_actions_memory_context	Read	133	1.1%
bighub_ingest_batch	Read	133	1.1%
bighub_outcomes_recommendation_quality	Read	133	1.1%
bighub_calibration_observe	Read	129	1.1%
bighub_kill_switch_activate	Write	127	1.1%
bighub_learning_recompute	Execute	119	1.0%
bighub_learning_backfill	Read	117	1.0%
bighub_constraints_purge_idempotency	Destructive	116	1.0%
bighub_ingest_stale	Read	115	1.0%
bighub_precedents_signals	Read	115	1.0%
bighub_kill_switch_deactivate	Write	114	1.0%
bighub_events_list	Read	113	1.0%
bighub_features_compare	Read	106	0.9%
bighub_constraints_list	Read	103	0.9%
bighub_insights_advise	Read	100	0.9%
bighub_outcomes_pending	Read	100	0.9%
bighub_webhooks_verify_signature	Read	100	0.9%
bighub_calibration_quality_history	Read	96	0.8%
bighub_actions_live_heartbeat	Read	94	0.8%
bighub_ingest_lifecycles	Read	93	0.8%
bighub_simulations_list	Read	93	0.8%
bighub_outcomes_partner_view	Read	92	0.8%
bighub_actions_dry_run	Execute	91	0.8%
bighub_webhooks_replay_failed_delivery	Read	91	0.8%
bighub_actions_live_connect	Write	88	0.7%
bighub_constraints_update	Write	88	0.7%
bighub_actions_evaluate_payload	Read	86	0.7%
bighub_actions_memory_refresh_aggregates	Read	86	0.7%
bighub_ingest_lifecycle	Read	86	0.7%
bighub_webhooks_deliveries	Read	84	0.7%
bighub_approvals_resolve	Write	84	0.7%
bighub_constraints_versions	Read	81	0.7%
bighub_actions_live_disconnect	Write	81	0.7%
bighub_features_export_batch	Write	81	0.7%
bighub_outcomes_report_batch	Read	80	0.7%
bighub_webhooks_update	Write	80	0.7%
bighub_approvals_list	Read	79	0.7%
bighub_calibration_drift	Read	77	0.7%
bighub_constraints_create	Write	75	0.6%
bighub_ingest_pending	Read	74	0.6%
bighub_constraints_dry_run	Execute	72	0.6%
bighub_insights_patterns	Read	72	0.6%
bighub_calibration_report	Read	71	0.6%
bighub_constraints_validate_dry_run	Read	71	0.6%
bighub_learning_runs	Read	71	0.6%
bighub_webhooks_test	Read	71	0.6%
bighub_auth_signup	Read	70	0.6%
bighub_auth_login	Write	69	0.6%
bighub_insights_profile	Read	67	0.6%
bighub_api_keys_create	Write	67	0.6%
bighub_webhooks_create	Write	67	0.6%
bighub_constraints_pause	Read	66	0.6%
bighub_constraints_validate	Read	66	0.6%
bighub_features_snapshot	Read	66	0.6%
bighub_features_explain	Read	65	0.6%
bighub_api_keys_delete	Destructive	64	0.5%
bighub_api_keys_validate	Read	64	0.5%
bighub_constraints_resume	Write	64	0.5%
bighub_constraints_delete	Destructive	63	0.5%
bighub_outcomes_analytics	Read	62	0.5%
bighub_simulations_by_request	Read	60	0.5%
bighub_simulations_compare	Read	60	0.5%
bighub_outcomes_timeline	Read	59	0.5%
bighub_webhooks_get	Read	59	0.5%
bighub_actions_verify_validation	Read	58	0.5%
bighub_outcomes_get_by_case	Read	58	0.5%
bighub_outcomes_get_by_validation	Read	58	0.5%
bighub_retrieval_strategy	Read	58	0.5%
bighub_calibration_reliability	Read	57	0.5%
bighub_constraints_get	Read	57	0.5%
bighub_features_get_snapshot	Read	57	0.5%
bighub_simulations_get	Read	57	0.5%
bighub_webhooks_delete	Destructive	56	0.5%
bighub_cases_get	Read	56	0.5%
bighub_features_list_snapshots	Read	56	0.5%
bighub_ingest_stats	Read	56	0.5%
bighub_outcomes_get	Read	56	0.5%
bighub_features_export	Write	56	0.5%
bighub_api_keys_rotate	Read	55	0.5%
bighub_cases_calibration	Read	54	0.5%
bighub_simulations_accuracy	Read	54	0.5%
bighub_api_keys_list	Read	53	0.5%
bighub_calibration_breakdown	Read	53	0.5%
bighub_webhooks_list	Read	53	0.5%
bighub_auth_refresh	Read	50	0.4%
bighub_features_schema	Read	50	0.4%
bighub_calibration_feedback	Read	49	0.4%
bighub_ingest_adapters	Read	49	0.4%
bighub_auth_logout	Read	48	0.4%
bighub_insights_learn	Execute	47	0.4%
bighub_retrieval_strategies	Read	47	0.4%
bighub_api_keys_scopes	Read	46	0.4%
bighub_learning_strategy	Read	46	0.4%
bighub_webhooks_list_events	Read	46	0.4%
bighub_actions_observer_stats	Read	45	0.4%
bighub_outcomes_taxonomy	Read	45	0.4%
bighub_precedents_stats	Read	45	0.4%
bighub_retrieval_stats	Read	45	0.4%
bighub_kill_switch_status	Write	45	0.4%
bighub_actions_dashboard_summary	Read	44	0.4%
bighub_simulations_stats	Read	44	0.4%
bighub_actions_status	Read	43	0.4%
bighub_constraints_domains	Read	43	0.4%
bighub_events_stats	Read	43	0.4%
bighub_features_stats	Read	43	0.4%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 125.

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 (94 tokens each).

Grant scope	Definition cost	Reduction
All 125 tools (no gateway)	11,747 tokens	—
3 granted tools	~282 tokens	−98%
5 granted tools	~470 tokens	−96%
10 granted tools	~940 tokens	−92%

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

//FAQ

BIGHUB token-cost questions.

How many tokens does the BIGHUB MCP server use?+

Its 125 tool definitions total 11,747 tokens — 5.9% 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 BIGHUB 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 BIGHUB's token usage?+

Expose fewer tools. A PolicyLayer grant scopes BIGHUB 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 282 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 125 catalogued BIGHUB tools. Counts refresh with every site build.