// MCP TOKEN COST

The Bitrise MCP server costs 11,259 tokens before the first call.

Connect Bitrise and its 81 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 Bitrise MCP server's tool definitions consume 11,259 tokens — 5.9× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 81 tools · 11,259 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: Bitrise ranks #147 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 11,259 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 139 tokens per tool.

Tool	Category	Tokens	% of server
list_installable_artifacts	Read	607	5.4%
generate_installable_artifact_upload_url	Write	401	3.6%
finish_bitrise_app	Execute	386	3.4%
create_connected_app	Write	378	3.4%
register_app	Write	357	3.2%
list_pipelines	Read	325	2.9%
codepush_generate_update_upload_url	Write	295	2.6%
update_connected_app	Write	279	2.5%
trigger_bitrise_build	Execute	267	2.4%
get_build_log	Read	257	2.3%
list_build_distribution_versions	Read	253	2.2%
codepush_promote_deployment	Write	247	2.2%
create_tester_group	Write	242	2.1%
list_connected_apps	Read	240	2.1%
list_builds	Read	226	2.0%
get_potential_testers	Read	222	2.0%
get_testers	Read	197	1.7%
codepush_patch_update	Write	178	1.6%
list_build_distribution_version_test_builds	Read	175	1.6%
update_bitrise_yml	Write	175	1.6%
step_search	Read	172	1.5%
update_tester_group	Write	162	1.4%
abort_build	Execute	159	1.4%
update_outgoing_webhook	Write	159	1.4%
validate_bitrise_yml	Read	155	1.4%
list_apps	Read	153	1.4%
list_tester_groups	Read	150	1.3%
set_installable_artifact_public_install_page	Write	146	1.3%
get_installable_artifact_upload_and_proc_status	Read	137	1.2%
codepush_list_updates	Read	136	1.2%
codepush_list_deployments	Read	134	1.2%
abort_pipeline	Read	131	1.2%
add_testers_to_tester_group	Write	131	1.2%
create_outgoing_webhook	Write	129	1.1%
notify_tester_group	Execute	126	1.1%
register_ssh_key	Write	119	1.1%
update_app	Write	110	1.0%
list_artifacts	Read	107	1.0%
update_artifact	Write	106	0.9%
rebuild_pipeline	Read	104	0.9%
get_bitrise_yml	Read	103	0.9%
list_cache_items	Read	103	0.9%
list_available_stacks	Read	96	0.9%
codepush_rollback_deployment	Read	95	0.8%
get_tester_group	Read	95	0.8%
codepush_create_deployment	Write	95	0.8%
replace_group_roles	Write	94	0.8%
list_outgoing_webhooks	Read	93	0.8%
get_build_steps	Read	89	0.8%
step_inputs	Read	87	0.8%
get_build	Read	85	0.8%
get_pipeline	Read	84	0.7%
get_artifact	Read	83	0.7%
delete_artifact	Destructive	82	0.7%
get_cache_item_download_url	Read	73	0.6%
get_build_bitrise_yml	Read	72	0.6%
delete_outgoing_webhook	Destructive	71	0.6%
codepush_update_deployment	Write	71	0.6%
delete_app	Destructive	69	0.6%
delete_cache_item	Destructive	69	0.6%
codepush_get_metrics	Read	69	0.6%
invite_member_to_workspace	Write	68	0.6%
create_workspace_group	Write	67	0.6%
list_group_roles	Read	66	0.6%
add_member_to_group	Write	65	0.6%
codepush_get_update_status	Read	64	0.6%
codepush_delete_deployment	Destructive	57	0.5%
codepush_delete_update	Destructive	56	0.5%
codepush_get_deployment	Read	54	0.5%
delete_all_cache_items	Destructive	53	0.5%
codepush_get_update	Read	53	0.5%
get_connected_app	Read	53	0.5%
list_branches	Read	53	0.5%
register_webhook	Write	50	0.4%
list_build_workflows	Read	49	0.4%
get_app	Read	48	0.4%
get_workspace_groups	Read	48	0.4%
get_workspace_members	Read	48	0.4%
get_workspace	Read	46	0.4%
list_workspaces	Read	26	0.2%
me	Read	24	0.2%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 81.

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 (139 tokens each).

Grant scope	Definition cost	Reduction
All 81 tools (no gateway)	11,259 tokens	—
3 granted tools	~417 tokens	−96%
5 granted tools	~695 tokens	−94%
10 granted tools	~1,390 tokens	−88%

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

//FAQ

Bitrise token-cost questions.

How many tokens does the Bitrise MCP server use?+

Its 81 tool definitions total 11,259 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 Bitrise 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 Bitrise's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Bitrise 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 417 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 81 catalogued Bitrise tools. Counts refresh with every site build.

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

A PolicyLayer grant scopes Bitrise 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.