Home / Token cost / Marketplace

The Marketplace MCP server costs 3,004 tokens before the first call.

Connect Marketplace and its 19 tool definitions are loaded into the model's context on every request — 1.5% of a 200k window spent before your agent does anything.

QUICK ANSWER The Marketplace MCP server's tool definitions consume 3,004 tokens — around the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 19 tools · 3,004 tokens · 1.5% of 200k · 0.3% of 1M Method →

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 1.5%
1M WINDOW 0.3%

Corpus context: Marketplace ranks #1302 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.

Where the 3,004 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 158 tokens per tool.

ToolCategoryTokens% of server
cars_session_tradein Execute 485 16.1%
cars_session_create Write 356 11.9%
cars_search Read 338 11.3%
cars_session_visit Read 252 8.4%
cars_feedback Execute 217 7.2%
cars_session_notify Write 174 5.8%
cars_session_message Read 160 5.3%
cars_session_add_car Write 135 4.5%
cars_session_reply Write 130 4.3%
cars_session_interest Read 119 4.0%
cars_session_read Read 102 3.4%
cars_dealers Read 91 3.0%
cars_session_tradein_read Read 82 2.7%
cars_history Read 75 2.5%
cars_dealer_inventory Read 66 2.2%
cars_session_compare Read 64 2.1%
cars_vehicle Read 62 2.1%
cars_session_deal Read 60 2.0%
cars_stats Read 36 1.2%

Most agents use a handful of these tools. They pay for all 19.

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 (158 tokens each).

Grant scopeDefinition costReduction
All 19 tools (no gateway) 3,004 tokens
3 granted tools ~474 tokens −84%
5 granted tools ~791 tokens −74%
10 granted tools ~1,581 tokens −47%

Marketplace token-cost questions.

How many tokens does the Marketplace MCP server use?+

Its 19 tool definitions total 3,004 tokens — 1.5% 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 Marketplace 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 Marketplace's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Marketplace 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 474 tokens, a 84% 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.

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 19 catalogued Marketplace tools. Counts refresh with every site build.

Expose only the tools you use — the rest never enter your context.

A PolicyLayer grant scopes Marketplace to the tools you actually allow. Ungranted definitions never load, and every call that does run is checked against policy first.

Free to start. No card required.

4,600+ MCP servers and 31,000+ tools scanned and risk-classified.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.