// MCP TOKEN COST

The MiOffice — AI-Powered Workspace Studio MCP server costs 16,574 tokens before the first call.

Connect MiOffice — AI-Powered Workspace Studio and its 133 tool definitions are loaded into the model's context on every request — 8.3% of a 200k window spent before your agent does anything.

QUICK ANSWER The MiOffice — AI-Powered Workspace Studio MCP server's tool definitions consume 16,574 tokens — 8.7× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 133 tools · 16,574 tokens · 8.3% of 200k · 1.7% 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 8.3%

1M WINDOW 1.7%

Corpus context: MiOffice — AI-Powered Workspace Studio ranks #77 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 16,574 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 125 tokens per tool.

Tool	Category	Tokens	% of server
mio_ai_music_generator	Destructive	165	1.0%
mio_ai_voice_generator	Destructive	157	0.9%
mio_ai_voice_cloner	Destructive	156	0.9%
mio_ai_melody_to_music	Destructive	152	0.9%
mio_ai_face_enhancer	Destructive	151	0.9%
mio_ai_upscale_pro	Destructive	150	0.9%
mio_ai_remove_background_pro	Destructive	149	0.9%
mio_ai_song_generator	Destructive	149	0.9%
mio_ai_audio_enhancer	Destructive	148	0.9%
mio_ai_inpaint_pro	Destructive	146	0.9%
mio_ai_vocal_remover	Destructive	146	0.9%
mio_ai_document_summarizer	Destructive	145	0.9%
mio_ai_photo_restorer	Destructive	144	0.9%
mio_ai_video_subtitler	Destructive	144	0.9%
mio_ai_silence_remover	Destructive	143	0.9%
mio_ai_talking_head	Destructive	143	0.9%
mio_video_resize_square	Write	143	0.9%
mio_ai_cartoon_filter	Destructive	142	0.9%
mio_ai_clip_maker	Destructive	142	0.9%
mio_ai_video_background_remover	Destructive	142	0.9%
mio_ai_video_enhancer	Destructive	142	0.9%
mio_video_auto_captions	Destructive	142	0.9%
mio_video_resize_reels	Write	142	0.9%
mio_ai_photo_colorizer	Destructive	141	0.9%
mio_ai_video_translator	Destructive	141	0.9%
mio_video_resize_shorts	Write	141	0.9%
mio_video_resize_tiktok	Write	141	0.9%
mio_ai_document_translator	Destructive	140	0.8%
mio_ai_headshot_generator	Destructive	140	0.8%
mio_ai_text_to_video	Destructive	140	0.8%
mio_ai_transcriber	Destructive	140	0.8%
mio_image_upscale	Destructive	139	0.8%
mio_ai_image_generator	Destructive	138	0.8%
mio_ai_logo_generator	Destructive	137	0.8%
mio_video_m4a_to_mp3	Execute	137	0.8%
mio_ai_face_swap	Destructive	136	0.8%
mio_ai_remove_background	Destructive	135	0.8%
mio_video_ogg_to_mp3	Execute	135	0.8%
mio_audio_converter	Execute	133	0.8%
mio_video_flac_to_mp3	Execute	133	0.8%
mio_video_mkv_to_mp4	Execute	133	0.8%
mio_video_webm_to_mp4	Execute	133	0.8%
mio_video_wma_to_mp3	Execute	133	0.8%
mio_video_mute	Destructive	132	0.8%
mio_video_wav_to_mp3	Execute	132	0.8%
mio_video_resize	Write	132	0.8%
mio_video_aac_to_mp3	Execute	131	0.8%
mio_video_avi_to_mp4	Execute	131	0.8%
mio_video_gif_to_mp4	Execute	131	0.8%
mio_video_extract_subtitles	Read	131	0.8%
mio_video_color	Execute	130	0.8%
mio_video_hdr_to_sdr	Execute	130	0.8%
mio_video_mov_to_mp4	Execute	130	0.8%
mio_video_rotate	Execute	130	0.8%
mio_video_add_audio	Write	130	0.8%
mio_video_denoise	Destructive	129	0.8%
mio_audio_equalizer	Execute	129	0.8%
mio_video_mp3_cutter	Execute	129	0.8%
mio_video_speed	Execute	129	0.8%
mio_video_thumbnail	Execute	129	0.8%
mio_video_converter	Execute	128	0.8%
mio_video_fade	Execute	128	0.8%
mio_audio_denoise	Destructive	127	0.8%
mio_audio_fade	Execute	127	0.8%
mio_audio_reverb	Execute	127	0.8%
mio_video_loop	Execute	127	0.8%
mio_video_to_gif	Execute	127	0.8%
mio_video_to_mp3	Execute	127	0.8%
mio_video_normalize_audio	Execute	126	0.8%
mio_video_trim	Execute	126	0.8%
mio_audio_compressor	Execute	125	0.8%
mio_audio_speed	Execute	125	0.8%
mio_video_reverse	Execute	124	0.7%
mio_video_merge	Write	124	0.7%
mio_video_compress	Execute	123	0.7%
mio_video_crop	Execute	123	0.7%
mio_video_flip	Execute	123	0.7%
mio_scanner_barcode	Other	121	0.7%
mio_p2p_session_handoff	Other	119	0.7%
mio_scanner_id_card	Other	118	0.7%
mio_pdf_unzip	Other	117	0.7%
mio_p2p_file_transfer	Financial	116	0.7%
mio_image_jpg_to_webp	Other	116	0.7%
mio_scanner_photo_to_pdf	Other	115	0.7%
mio_scanner_whiteboard	Other	115	0.7%
mio_image_heic_to_jpg	Other	114	0.7%
mio_image_png_to_webp	Other	114	0.7%
mio_p2p_screen_share	Other	114	0.7%
mio_image_heic_to_png	Other	113	0.7%
mio_pdf_from_word	Other	113	0.7%
mio_scanner_handwriting	Other	113	0.7%
mio_image_avif_to_jpg	Other	112	0.7%
mio_image_webp_to_jpg	Other	112	0.7%
mio_scanner_batch	Other	112	0.7%
mio_scanner_qr	Other	112	0.7%
mio_scanner_receipt	Other	112	0.7%
mio_ai_remove_object	Destructive	111	0.7%
mio_image_webp_to_png	Other	111	0.7%
mio_scanner_book	Other	111	0.7%
mio_scanner_document	Other	111	0.7%
mio_pdf_editor	Write	111	0.7%
mio_image_jpeg_to_jpg	Other	110	0.7%
mio_jpg_to_pdf	Other	110	0.7%
mio_notes_notes	Other	110	0.7%
mio_pdf_epub_to_pdf	Other	110	0.7%
mio_pdf_tiff_to_pdf	Other	110	0.7%
mio_pdf_to_jpg	Other	110	0.7%
mio_image_compress_webp	Other	109	0.7%
mio_image_png_to_svg	Other	109	0.7%
mio_pdf_split	Other	109	0.7%
mio_pdf_to_xlsx	Other	109	0.7%
mio_pdf_remove_pages	Destructive	108	0.7%
mio_pdf_compress	Other	108	0.7%
mio_pdf_image_to_pdf	Other	108	0.7%
mio_pdf_png_to_pdf	Other	108	0.7%
mio_pdf_to_doc	Other	108	0.7%
mio_pdf_txt_to_pdf	Other	108	0.7%
mio_xlsx_to_pdf	Other	108	0.7%
mio_image_compress	Other	107	0.6%
mio_pdf_to_text	Other	107	0.6%
mio_pdf_watermark	Write	107	0.6%
mioffice_open_tool	Write	107	0.6%
mio_image_crop	Other	106	0.6%
mio_pdf_merge	Write	106	0.6%
mio_pdf_rotate	Other	105	0.6%
mio_image_resize	Write	105	0.6%
mio_pdf_unlock	Write	105	0.6%
mio_pdf_zip	Write	105	0.6%
mioffice_list_tools	Read	104	0.6%
mio_image_rotate	Other	103	0.6%
mio_image_convert	Write	103	0.6%
mioffice_pricing_info	Other	73	0.4%
mioffice_pdf_editor	Other	65	0.4%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 133.

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 (125 tokens each).

Grant scope	Definition cost	Reduction
All 133 tools (no gateway)	16,574 tokens	—
3 granted tools	~374 tokens	−98%
5 granted tools	~623 tokens	−96%
10 granted tools	~1,246 tokens	−92%

Model your own stack in the token-cost calculator, or see the MiOffice — AI-Powered Workspace Studio policy for what a sensible grant looks like.

//FAQ

MiOffice — AI-Powered Workspace Studio token-cost questions.

How many tokens does the MiOffice — AI-Powered Workspace Studio MCP server use?+

Its 133 tool definitions total 16,574 tokens — 8.3% 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 MiOffice — AI-Powered Workspace Studio 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 MiOffice — AI-Powered Workspace Studio's token usage?+

Expose fewer tools. A PolicyLayer grant scopes MiOffice — AI-Powered Workspace Studio 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 374 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 133 catalogued MiOffice — AI-Powered Workspace Studio tools. Counts refresh with every site build.