// MCP TOKEN COST

The Propresenter MCP server costs 8,087 tokens before the first call.

Connect Propresenter and its 177 tool definitions are loaded into the model's context on every request — 4.0% of a 200k window spent before your agent does anything.

QUICK ANSWER The Propresenter MCP server's tool definitions consume 8,087 tokens — 4.2× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 177 tools · 8,087 tokens · 4.0% of 200k · 0.8% 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 4.0%

1M WINDOW 0.8%

Corpus context: Propresenter ranks #269 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 8,087 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 46 tokens per tool.

Tool	Category	Tokens	% of server
library_trigger_presentation_cue	Execute	96	1.2%
timers_operation	Destructive	87	1.1%
get_presentation_thumbnail	Read	87	1.1%
themes_set_slide	Write	85	1.1%
playlists_get_thumbnail	Read	84	1.0%
playlists_create_under	Write	83	1.0%
library_trigger_presentation	Execute	79	1.0%
audio_get_playlist	Read	78	1.0%
presentation_timeline_operation	Read	78	1.0%
trigger_presentation	Execute	77	1.0%
trigger_presentation_cue	Execute	77	1.0%
transport_skip_backward	Read	77	1.0%
transport_skip_forward	Read	77	1.0%
clear_layer	Destructive	76	0.9%
transport_set_time	Write	74	0.9%
transport_go_to_end	Read	73	0.9%
trigger_presentation_group	Execute	72	0.9%
audio_trigger_playlist_item	Execute	71	0.9%
messages_trigger	Execute	71	0.9%
playlists_create	Write	70	0.9%
stage_set_screen_layout	Write	70	0.9%
playlists_get_active_announcement_thumbnail	Read	68	0.8%
playlists_get_active_presentation_thumbnail	Read	68	0.8%
themes_get_slide_thumbnail	Read	67	0.8%
timers_increment	Read	67	0.8%
playlists_trigger_item	Execute	66	0.8%
themes_get_slide	Read	66	0.8%
timers_operate_all	Read	66	0.8%
playlists_set	Write	65	0.8%
announcement_timeline_operation	Read	63	0.8%
audio_get_playlist_updates	Read	63	0.8%
focused_timeline_operation	Read	63	0.8%
capture_get_encodings	Read	62	0.8%
transport_get_auto_advance	Read	62	0.8%
messages_update	Write	62	0.8%
timeline_operation	Read	61	0.8%
props_set_collection	Write	61	0.8%
transport_cancel_auto_advance	Destructive	60	0.7%
props_set	Write	59	0.7%
transport_get_current	Read	58	0.7%
transport_get_time	Read	58	0.7%
transport_pause	Read	58	0.7%
transport_play	Read	58	0.7%
trigger_focused_cue	Execute	57	0.7%
announcement_trigger_cue	Execute	56	0.7%
clear_get_group	Destructive	55	0.7%
library_get	Read	55	0.7%
capture_operation	Execute	54	0.7%
looks_get	Read	53	0.7%
macros_get	Read	53	0.7%
playlists_trigger_active_announcement_item	Execute	52	0.6%
playlists_trigger_active_presentation_item	Execute	52	0.6%
trigger_focused_group	Execute	52	0.6%
trigger_presentation_next	Execute	52	0.6%
trigger_presentation_previous	Execute	52	0.6%
messages_create	Write	52	0.6%
messages_clear	Destructive	51	0.6%
trigger_active_group	Execute	51	0.6%
looks_trigger	Execute	50	0.6%
playlists_trigger_focused_item	Execute	50	0.6%
playlists_trigger_next	Execute	50	0.6%
playlists_trigger_previous	Execute	50	0.6%
playlists_focus	Read	50	0.6%
focus_presentation	Read	49	0.6%
get_presentation	Read	49	0.6%
stage_get_screen_layout	Read	49	0.6%
stage_set_layout_map	Write	49	0.6%
audio_trigger_playlist	Execute	48	0.6%
audio_focus_playlist	Read	48	0.6%
playlists_get_updates	Read	48	0.6%
clear_trigger_group	Destructive	47	0.6%
props_pause_auto_clear	Destructive	47	0.6%
timers_reset	Destructive	47	0.6%
playlists_trigger	Execute	47	0.6%
timers_start	Execute	47	0.6%
timers_stop	Execute	47	0.6%
playlists_get	Read	47	0.6%
stage_get_layout_thumbnail	Read	47	0.6%
timers_get	Read	47	0.6%
props_resume_auto_clear	Write	47	0.6%
messages_delete	Destructive	46	0.6%
props_clear	Destructive	46	0.6%
props_delete_collection	Destructive	46	0.6%
stage_delete_layout	Destructive	46	0.6%
props_trigger	Execute	46	0.6%
masks_get_thumbnail	Read	46	0.6%
messages_get	Read	46	0.6%
props_get_collection	Read	46	0.6%
status_set_audience	Read	46	0.6%
macros_trigger	Execute	45	0.6%
masks_get	Read	45	0.6%
props_get_thumbnail	Read	45	0.6%
stage_show_message	Read	45	0.6%
status_post_updates	Read	45	0.6%
status_set_stage	Read	45	0.6%
props_create_collection	Write	45	0.6%
props_delete	Destructive	44	0.5%
props_get	Read	44	0.5%
themes_get	Read	44	0.5%
looks_get_all	Read	36	0.4%
find_my_mouse	Read	35	0.4%
trigger_active_presentation	Execute	34	0.4%
announcement_get_slide_index	Read	34	0.4%
get_active_slide_index	Read	34	0.4%
get_focused_timeline_status	Read	34	0.4%
announcement_trigger_active	Execute	33	0.4%
audio_trigger_active_playlist	Execute	33	0.4%
playlists_trigger_focused_next	Execute	33	0.4%
playlists_trigger_focused_previous	Execute	33	0.4%
announcement_get_timeline_status	Read	33	0.4%
playlists_get_focused_next	Read	33	0.4%
playlists_get_focused_previous	Read	33	0.4%
status_get_screens	Read	33	0.4%
audio_trigger_focused_next	Execute	32	0.4%
audio_trigger_focused_previous	Execute	32	0.4%
playlists_trigger_active_announcement	Execute	32	0.4%
playlists_trigger_active_presentation	Execute	32	0.4%
get_propresenter_version	Read	32	0.4%
get_timeline_status	Read	32	0.4%
timers_get_system_time	Read	32	0.4%
audio_trigger_active_next	Execute	31	0.4%
audio_trigger_active_previous	Execute	31	0.4%
trigger_focused_next	Execute	31	0.4%
trigger_focused_previous	Execute	31	0.4%
audio_get_playlists	Read	31	0.4%
global_groups_get_all	Read	31	0.4%
looks_get_current	Read	31	0.4%
timers_get_video_countdown	Read	31	0.4%
clear_get_groups	Destructive	30	0.4%
announcement_trigger_next	Execute	30	0.4%
announcement_trigger_previous	Execute	30	0.4%
trigger_next_slide	Execute	30	0.4%
trigger_previous_slide	Execute	30	0.4%
audio_get_focused_playlist	Read	30	0.4%
capture_get_status	Read	30	0.4%
library_get_libraries	Read	30	0.4%
macros_get_all	Read	30	0.4%
playlists_focus_active_announcement	Read	30	0.4%
playlists_focus_active_presentation	Read	30	0.4%
stage_get_layout_map	Read	30	0.4%
stage_get_screens	Read	30	0.4%
timers_get_all	Read	30	0.4%
audio_trigger_focused_playlist	Execute	29	0.4%
playlists_trigger_focused	Execute	29	0.4%
trigger_focused_presentation	Execute	29	0.4%
audio_get_active_playlist	Read	29	0.4%
get_chord_chart	Read	29	0.4%
get_focused_presentation	Read	29	0.4%
masks_get_all	Read	29	0.4%
playlists_get_all	Read	29	0.4%
playlists_get_focused	Read	29	0.4%
status_get_audience	Read	29	0.4%
timers_get_current	Read	29	0.4%
announcement_focus_active	Read	28	0.3%
announcement_get_active	Read	28	0.3%
audio_focus_active_playlist	Read	28	0.3%
audio_focus_next_playlist	Read	28	0.3%
audio_focus_previous_playlist	Read	28	0.3%
focus_active_presentation	Read	28	0.3%
get_active_presentation	Read	28	0.3%
get_chord_chart_updates	Read	28	0.3%
messages_get_all	Read	28	0.3%
playlists_get_active	Read	28	0.3%
props_get_all	Read	28	0.3%
stage_get_layouts	Read	28	0.3%
status_get_layers	Read	28	0.3%
status_get_slide	Read	28	0.3%
status_get_stage	Read	28	0.3%
themes_get_all	Read	28	0.3%
capture_get_settings	Read	27	0.3%
focus_next_presentation	Read	27	0.3%
focus_previous_presentation	Read	27	0.3%
playlists_focus_next	Read	27	0.3%
playlists_focus_previous	Read	27	0.3%
props_get_collections	Read	27	0.3%
stage_get_message	Read	27	0.3%
stage_hide_message	Read	27	0.3%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 177.

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 (46 tokens each).

Grant scope	Definition cost	Reduction
All 177 tools (no gateway)	8,087 tokens	—
3 granted tools	~137 tokens	−98%
5 granted tools	~228 tokens	−97%
10 granted tools	~457 tokens	−94%

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

//FAQ

Propresenter token-cost questions.

How many tokens does the Propresenter MCP server use?+

Its 177 tool definitions total 8,087 tokens — 4.0% 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 Propresenter 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 Propresenter's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Propresenter 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 137 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 177 catalogued Propresenter tools. Counts refresh with every site build.