// MCP TOKEN COST

The Freedom MCP server costs 57,931 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 Freedom Mcp MCP server's 238 tool definitions consume 57,931 tokens — 29% of a 200k context window, and 29× the median MCP server (1,990 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS tiktoken o200k_base · rank #21 of 6,595 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 29%

1M WINDOW 5.8%

Corpus context: Freedom ranks #21 of 6,595 measured MCP servers by definition cost. The median is 1,990 tokens, p90 is 11,763, and the heaviest (Ainumbers Mcp Apps) is 255,699 — 128% 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 57,931 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 243 tokens per tool.

Tool	Category	Tokens	% of server
draft_outreach	Write	932	1.6%
submit_product_request	Write	861	1.5%
hire_agent_with_context	Execute	798	1.4%
configure_dashboard	Write	775	1.3%
send_email	Execute	717	1.2%
generate_carousel	Write	705	1.2%
generate_video	Write	690	1.2%
update_lead	Write	666	1.1%
update_my_profile	Write	596	1.0%
generate_image_xai	Execute	563	1.0%
update_agent_activity	Write	541	0.9%
create_icp	Write	533	0.9%
add_agent_activity	Write	530	0.9%
update_key_result	Write	528	0.9%
create_meta_ad_draft	Write	503	0.9%
save_knowledge	Write	501	0.9%
report_feedback	Write	482	0.8%
add_customer_evidence	Write	453	0.8%
deliberate	Execute	445	0.8%
update_agent_avatar	Write	423	0.7%
update_agent_skill	Write	411	0.7%
create_attention_directive	Write	405	0.7%
decide_command_center_item	Execute	403	0.7%
update_tactic	Write	403	0.7%
update_projection	Write	400	0.7%
generate_vector_image	Execute	395	0.7%
vectorize_image	Write	390	0.7%
create_key_result	Write	384	0.7%
create_pipeline	Write	376	0.6%
query_sme	Read	373	0.6%
upsert_attention_session	Write	370	0.6%
generate_video_veo	Execute	362	0.6%
set_revenue_channels	Write	355	0.6%
update_pipeline	Write	348	0.6%
get_release_ledger	Read	344	0.6%
create_tactic	Write	343	0.6%
set_grain_policy	Write	342	0.6%
confirm_mcp_approval	Write	341	0.6%
request_attention_spawn	Execute	338	0.6%
get_ads_performance	Read	337	0.6%
add_lead	Write	336	0.6%
generate_tactics	Write	328	0.6%
create_feature	Write	323	0.6%
publish_pipeline_item	Write	323	0.6%
ingest_voice_corpus	Write	320	0.6%
run_tactic	Execute	319	0.6%
ratify_capability	Write	314	0.5%
submit_content_to_pipeline	Write	312	0.5%
segment_leads	Read	308	0.5%
update_feature	Write	305	0.5%
get_search_performance	Read	304	0.5%
manage_responsibilities	Write	302	0.5%
browse_url	Execute	299	0.5%
posthog_query_trends	Read	295	0.5%
set_attention_budget	Write	294	0.5%
set_offer	Write	294	0.5%
add_commitment	Write	293	0.5%
get_decision_ledger	Read	292	0.5%
propose_work	Write	292	0.5%
update_commitment	Write	290	0.5%
scan_product_signals	Read	286	0.5%
audit_brand_visibility	Execute	276	0.5%
remove_background	Destructive	273	0.5%
check_my_inbox	Read	273	0.5%
get_check_telemetry	Read	272	0.5%
send_slack_message	Write	272	0.5%
synthesize_lead_hypothesis	Write	272	0.5%
get_command_center_items	Read	271	0.5%
grant_agent_tool	Write	270	0.5%
update_reader_profile	Write	270	0.5%
get_next_priority	Read	268	0.5%
create_play_from_activity	Write	267	0.5%
split_agent_activity	Write	265	0.5%
set_cac_strategy	Write	265	0.5%
update_objective	Write	265	0.5%
update_knowledge_section	Write	263	0.5%
remove_agent_activity	Destructive	261	0.5%
run_quality_check	Execute	259	0.4%
update_agent	Write	258	0.4%
trigger_agent_activity	Execute	256	0.4%
delete_key_result	Write	255	0.4%
save_artifact	Write	254	0.4%
share_knowledge	Write	254	0.4%
promote_corpus_to_content	Write	250	0.4%
challenge_as_customer	Execute	248	0.4%
request_connector	Write	248	0.4%
link_agent_okrs	Write	247	0.4%
recalibrate_agent_jd	Write	246	0.4%
derive_from_website	Read	244	0.4%
get_xero_report	Read	242	0.4%
send_lead_draft	Write	242	0.4%
retire_feature	Write	240	0.4%
archive_pipeline	Write	240	0.4%
read_knowledge	Read	238	0.4%
derive_capability	Read	233	0.4%
update_pipeline_style_guide	Write	232	0.4%
toggle_agent_schedule	Execute	230	0.4%
posthog_query_funnel	Read	229	0.4%
create_objective	Write	229	0.4%
revoke_agent_tool	Write	228	0.4%
draft_tenet_from_signal	Read	228	0.4%
claim_product_request_for_builder	Write	227	0.4%
create_google_doc	Write	225	0.4%
resolve_work	Execute	224	0.4%
suggest_collaboration	Write	224	0.4%
set_meta_ad_status	Financial	218	0.4%
approve_pipeline_item	Write	215	0.4%
get_page_performance	Read	214	0.4%
resolve_brand_guide	Read	213	0.4%
update_meta_ad_budget	Financial	213	0.4%
delete_icp	Destructive	212	0.4%
get_partner_cos_onboard	Read	211	0.4%
update_icp	Write	208	0.4%
deactivate_agent	Destructive	205	0.4%
append_cos_preference	Write	205	0.4%
triage_idea	Write	205	0.4%
list_my_work	Read	204	0.4%
attach_product_request_pr	Write	203	0.4%
get_transactions	Read	202	0.3%
request_content_revision	Write	200	0.3%
get_cac_strategy	Read	199	0.3%
update_company	Write	199	0.3%
cancel_commitment	Write	198	0.3%
originate_content_ideas	Write	198	0.3%
update_finance_note	Write	198	0.3%
sync_stripe_conversions	Write	195	0.3%
list_ad_campaigns	Read	194	0.3%
open_product_request_draft_pr	Write	194	0.3%
share_commitment	Write	194	0.3%
reassign_reports	Write	193	0.3%
get_setup_state	Read	192	0.3%
update_brand_guidelines	Write	191	0.3%
get_icps	Read	190	0.3%
list_attention_sessions	Read	189	0.3%
generate_key_results	Write	189	0.3%
clear_pipeline_learnings	Destructive	188	0.3%
delete_objective	Destructive	188	0.3%
get_financial_summary	Read	187	0.3%
get_my_channel_partner_stats	Read	187	0.3%
invoke_integration	Execute	184	0.3%
update_feature_status	Write	184	0.3%
list_customer_evidence	Read	182	0.3%
hire_agent	Execute	178	0.3%
update_sheet	Write	178	0.3%
update_voice_profile	Write	178	0.3%
get_attention_budget	Read	177	0.3%
inspect_url	Read	176	0.3%
ack_attention_directive	Write	176	0.3%
get_agent_outcome_panel	Read	175	0.3%
interview_for_hire	Write	174	0.3%
list_pipeline_learnings	Read	173	0.3%
get_artifacts	Read	172	0.3%
get_x_post_metrics	Read	172	0.3%
get_product_context	Read	171	0.3%
list_deals	Read	170	0.3%
list_xero_bank_transactions	Read	170	0.3%
create_master_plan	Write	169	0.3%
unshare_commitment	Write	169	0.3%
batch_update_spreadsheet	Write	168	0.3%
get_pending_approvals	Read	167	0.3%
list_segments	Read	167	0.3%
search_conversations	Read	167	0.3%
get_lead_pipeline_snapshot	Read	166	0.3%
get_top_customers	Read	166	0.3%
list_pipelines	Read	166	0.3%
search_connector_registry	Read	166	0.3%
get_brand_guidelines	Read	165	0.3%
create_folder	Write	165	0.3%
list_xero_contacts	Read	164	0.3%
get_team_pulse	Read	163	0.3%
read_web_page	Read	163	0.3%
get_credit_usage	Read	162	0.3%
cancel_attention_directive	Write	161	0.3%
delete_tactic	Write	160	0.3%
get_my_channel_partner_starter_pack	Read	158	0.3%
update_transaction_note	Write	158	0.3%
unshare_knowledge	Write	157	0.3%
get_my_channel_partner_link	Read	155	0.3%
set_cos_preferences	Write	154	0.3%
get_stripe_metrics	Read	152	0.3%
get_okrs	Read	151	0.3%
complete_commitment	Write	151	0.3%
delete_idea	Destructive	150	0.3%
delete_knowledge	Destructive	150	0.3%
get_command_center_item	Read	150	0.3%
get_reader_expertise_interview	Read	149	0.3%
list_knowledge	Read	148	0.3%
append_to_sheet	Write	147	0.3%
capture_idea	Write	147	0.3%
get_agent_performance	Read	146	0.3%
query_lead_journey	Read	144	0.2%
get_tactics	Read	143	0.2%
list_attention_directives	Read	143	0.2%
analyze_team_needs	Read	142	0.2%
list_google_drive_files	Read	142	0.2%
get_reader_profile	Read	141	0.2%
get_voice_profile	Read	141	0.2%
get_grain_policy	Read	140	0.2%
create_spreadsheet	Write	140	0.2%
get_my_companies	Read	139	0.2%
update_google_doc	Write	139	0.2%
search_ad_targeting	Read	138	0.2%
posthog_list_events	Read	136	0.2%
get_sitemaps	Read	135	0.2%
get_activity_health	Read	133	0.2%
list_integrations	Read	132	0.2%
get_product_request_status	Read	131	0.2%
list_ad_accounts	Read	130	0.2%
posthog_hogql	Execute	129	0.2%
list_features	Read	129	0.2%
preview_meta_ad	Read	128	0.2%
list_inbox	Read	126	0.2%
list_commitments	Read	123	0.2%
list_corpus_inventory	Read	123	0.2%
list_workspace_ideas	Read	123	0.2%
remove_dashboard_widget	Destructive	120	0.2%
get_executive_landscape	Read	120	0.2%
posthog_list_insights	Read	118	0.2%
list_scheduled_reports	Read	116	0.2%
get_monthly_trends	Read	113	0.2%
get_actuals_vs_budget	Read	111	0.2%
get_projections	Read	111	0.2%
read_sheet	Read	111	0.2%
search_transactions	Read	110	0.2%
suggest_next_hire	Read	107	0.2%
list_dashboard_widgets	Read	101	0.2%
get_subscription_stats	Read	99	0.2%
get_freedom_target	Read	96	0.2%
get_team_members	Read	94	0.2%
read_google_doc	Read	94	0.2%
get_company	Read	90	0.2%
get_cos_preferences	Read	84	0.1%
get_team_roster	Read	83	0.1%
list_shared_with_me	Read	83	0.1%
get_site_list	Read	81	0.1%
get_routing_overview	Read	76	0.1%
get_my_profile	Read	71	0.1%
get_cash_position	Read	70	0.1%

// CUT THE COST

Your agent uses a handful of these tools. It pays for all 238.

You don't need all 238 of those definitions in the window. PolicyLayer is an MCP gateway that sits in front of Freedom Mcp: 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 238 tools (no gateway)	57,931 tokens	—
3 granted tools	~730 tokens	−99%
5 granted tools	~1,217 tokens	−98%
10 granted tools	~2,434 tokens	−96%

The risk dividend: 11 of these 238 tools are critical-risk (destructive or financial) and cost 2,178 tokens (4% 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 Freedom Mcp — 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 FREEDOM TOKEN COST →

Instant setup, no code required.

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

//FAQ

Freedom Mcp token-cost questions.

How many tokens does the Freedom MCP server use?+

Its 238 tool definitions total 57,931 tokens — 29% 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 Freedom Mcp 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 Freedom Mcp's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Freedom Mcp 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 730 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 08-08-2026 from the PolicyLayer scan database over all 238 catalogued Freedom Mcp tools. Counts refresh with every site build.