// MCP TOKEN COST

The ChiefLab MCP server costs 11,133 tokens before the first call.

Connect ChiefLab and its 45 tool definitions are loaded into the model's context on every request — 5.6% of a 200k window spent before your agent does anything.

QUICK ANSWER The ChiefLab MCP server's tool definitions consume 11,133 tokens — 5.8× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 45 tools · 11,133 tokens · 5.6% of 200k · 1.1% 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 5.6%

1M WINDOW 1.1%

Corpus context: ChiefLab ranks #151 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 11,133 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 247 tokens per tool.

Tool	Category	Tokens	% of server
chieflab_get_users_after_build	Read	586	5.3%
chieflab_approve_action	Write	527	4.7%
chieflab_use_manual_fallback	Write	497	4.5%
chieflab_execute_approved_action	Execute	472	4.2%
chieflab_regenerate_visual_asset	Execute	438	3.9%
chieflab_record_manual_metrics	Read	349	3.1%
chieflab_request_specialist	Read	338	3.0%
chieflab_prepare_next_move	Write	327	2.9%
chieflab_suggest_next_move	Write	327	2.9%
chieflab_publish_approved_post	Write	322	2.9%
chieflab_create_work_request	Write	319	2.9%
chieflab_record_manual_publish	Write	313	2.8%
chieflab_send_approved_email	Write	282	2.5%
chieflab_post_launch_review	Write	281	2.5%
chieflab_status	Read	260	2.3%
chieflab_review_visual_asset	Execute	259	2.3%
chieflab_create_next_move_action	Write	253	2.3%
chieflab_signup_workspace	Execute	249	2.2%
chieflab_continue_launch_loop	Execute	246	2.2%
chieflab_set_company_profile	Write	245	2.2%
chieflab_set_publishing_targets	Write	245	2.2%
chieflab_query_proof_ledger	Read	241	2.2%
chieflab_force_measurement_due	Destructive	236	2.1%
chieflab_redraft	Write	231	2.1%
chieflab_launch_product	Execute	218	2.0%
chieflab_measure_reddit	Write	216	1.9%
chieflab_measure_hacker_news	Write	212	1.9%
chieflab_set_provider_key	Write	211	1.9%
chieflab_verify_install	Read	207	1.9%
chieflab_mark_action_done	Write	200	1.8%
chieflab_check_measurement_due	Read	191	1.7%
chieflab_inbox	Read	188	1.7%
chieflab_search_capabilities	Read	173	1.6%
chieflab_connect_provider	Write	171	1.5%
chieflab_persistence_health	Execute	168	1.5%
chieflab_skip_next_move	Write	167	1.5%
chieflab_boot	Write	164	1.5%
chieflab_brain_summary	Read	136	1.2%
chieflab_retry_blocked_action	Write	131	1.2%
chieflab_help	Execute	100	0.9%
chieflab_connector_status	Read	94	0.8%
chieflab_get_publishing_targets	Read	94	0.8%
chieflab_list_specialists	Read	91	0.8%
chieflab_get_capability	Read	88	0.8%
chieflab_get_company_profile	Read	70	0.6%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 45.

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 (247 tokens each).

Grant scope	Definition cost	Reduction
All 45 tools (no gateway)	11,133 tokens	—
3 granted tools	~742 tokens	−93%
5 granted tools	~1,237 tokens	−89%
10 granted tools	~2,474 tokens	−78%

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

//FAQ

ChiefLab token-cost questions.

How many tokens does the ChiefLab MCP server use?+

Its 45 tool definitions total 11,133 tokens — 5.6% 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 ChiefLab 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 ChiefLab's token usage?+

Expose fewer tools. A PolicyLayer grant scopes ChiefLab 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 742 tokens, a 93% 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 45 catalogued ChiefLab tools. Counts refresh with every site build.

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

A PolicyLayer grant scopes ChiefLab 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.