// MCP TOKEN COST

The Onyx MCP server costs 12,920 tokens before the first call.

Connect Onyx and its 67 tool definitions are loaded into the model's context on every request — 6.5% of a 200k window spent before your agent does anything.

QUICK ANSWER The Onyx Mcp MCP server's tool definitions consume 12,920 tokens — 6.8× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 67 tools · 12,920 tokens · 6.5% of 200k · 1.3% 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 6.5%

1M WINDOW 1.3%

Corpus context: Onyx ranks #114 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 12,920 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 193 tokens per tool.

Tool	Category	Tokens	% of server
onyx_base_bridge_quote	Read	334	2.6%
onyx_mcp_router	Read	333	2.6%
onyx_research_intel	Read	301	2.3%
onyx_base_event_logs	Read	292	2.3%
onyx_x402_receipt_verify	Read	285	2.2%
onyx_bazaar_compare	Read	282	2.2%
onyx_verify_explain	Read	273	2.1%
onyx_x402_simulate	Read	268	2.1%
onyx_solana_jupiter_quote	Read	265	2.1%
onyx_base_contract_verify	Read	256	2.0%
onyx_base_tx_simulator	Read	251	1.9%
onyx_agent_audit_trail	Read	242	1.9%
onyx_base_swap_quote	Read	237	1.8%
onyx_x402_error_explain	Read	233	1.8%
onyx_bazaar_blue_ocean	Execute	228	1.8%
onyx_mcp_meta_call	Read	227	1.8%
onyx_skill_bundle	Read	222	1.7%
onyx_aml_screen	Read	220	1.7%
onyx_solve_captcha	Read	217	1.7%
onyx_base_token_risk_scan	Read	213	1.6%
onyx_base_tx_explainer	Read	211	1.6%
onyx_fx_convert	Write	211	1.6%
onyx_agent_workflow	Execute	210	1.6%
onyx_fact_check	Read	204	1.6%
onyx_x402_chain_picker	Read	203	1.6%
onyx_mcp_health	Read	198	1.5%
onyx_solana_wallet_activity	Read	197	1.5%
onyx_solana_token_risk_scan	Read	195	1.5%
onyx_agent_budget_tracker	Read	194	1.5%
onyx_x402_facilitators	Read	192	1.5%
onyx_pm_settlement_watch	Read	189	1.5%
onyx_solana_tx_explainer	Read	189	1.5%
onyx_mcp_registry_status	Read	187	1.4%
onyx_agent_id	Read	181	1.4%
onyx_base_dex_pair_lookup	Read	180	1.4%
onyx_partnership_check	Read	177	1.4%
onyx_x402_spec_lookup	Read	175	1.4%
onyx_x402_demo_wallet	Read	173	1.3%
onyx_browser_navigate	Execute	171	1.3%
onyx_browser_screenshot	Read	169	1.3%
onyx_facilitator_health	Read	168	1.3%
onyx_ip_geolocate	Read	166	1.3%
onyx_paper_synthesis	Read	166	1.3%
onyx_url_text	Read	166	1.3%
onyx_mcp_catalog_diff	Read	162	1.3%
onyx_x402_indexer_health	Read	157	1.2%
onyx_browser_type	Write	156	1.2%
onyx_solana_token_metadata	Read	153	1.2%
onyx_robots_check	Read	152	1.2%
onyx_base_tx_decode	Read	151	1.2%
onyx_ens_resolve	Write	150	1.2%
onyx_jwt_decode	Read	147	1.1%
onyx_bazaar_submit	Write	147	1.1%
onyx_dns_lookup	Read	146	1.1%
onyx_password_strength	Execute	145	1.1%
onyx_token_metadata	Read	145	1.1%
onyx_browser_eval	Read	144	1.1%
onyx_hash_compute	Read	142	1.1%
onyx_email_validate	Read	141	1.1%
onyx_browser_extract	Read	139	1.1%
onyx_html_meta	Execute	138	1.1%
onyx_whois	Read	134	1.0%
onyx_mcp_oauth_audit	Read	133	1.0%
onyx_url_unshorten	Read	133	1.0%
onyx_url_parse	Execute	122	0.9%
onyx_user_agent_parse	Execute	121	0.9%
onyx_browser_click	Write	111	0.9%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 67.

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 (193 tokens each).

Grant scope	Definition cost	Reduction
All 67 tools (no gateway)	12,920 tokens	—
3 granted tools	~579 tokens	−96%
5 granted tools	~964 tokens	−93%
10 granted tools	~1,928 tokens	−85%

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

//FAQ

Onyx Mcp token-cost questions.

How many tokens does the Onyx MCP server use?+

Its 67 tool definitions total 12,920 tokens — 6.5% 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 Onyx 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 Onyx Mcp's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Onyx 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 579 tokens, a 96% 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.

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.

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.

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 67 catalogued Onyx Mcp tools. Counts refresh with every site build.

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

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

SCOPE YOUR GRANT →

Free to start. No card required.

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