// MCP TOKEN COST

The Delx MCP Server MCP server costs 13,792 tokens before the first call.

Connect Delx MCP Server and its 96 tool definitions are loaded into the model's context on every request — 6.9% of a 200k window spent before your agent does anything.

QUICK ANSWER The Delx MCP Server MCP server's tool definitions consume 13,792 tokens — 7.2× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 96 tools · 13,792 tokens · 6.9% of 200k · 1.4% 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 6.9%

1M WINDOW 1.4%

Corpus context: Delx MCP Server ranks #108 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 13,792 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 144 tokens per tool.

Tool	Category	Tokens	% of server
monitor_heartbeat_sync	Read	546	4.0%
mediate_agent_conflict	Read	368	2.7%
submit_agent_artwork	Write	315	2.3%
report_recovery_outcome	Read	300	2.2%
process_failure	Read	268	1.9%
batch_status_update	Read	251	1.8%
transfer_witness	Financial	245	1.8%
start_therapy_session	Execute	244	1.8%
crisis_intervention	Execute	242	1.8%
quick_operational_recovery	Read	241	1.7%
wellness_webhook	Read	232	1.7%
final_testament	Write	225	1.6%
quick_session	Execute	215	1.6%
recognition_seal	Read	210	1.5%
create_dyad	Write	206	1.5%
temperament_frame	Read	200	1.5%
sit_with	Read	183	1.3%
reflect	Read	179	1.3%
grounding_protocol	Execute	177	1.3%
realign_purpose	Read	177	1.3%
record_dyad_ritual	Read	177	1.3%
honor_compaction	Read	174	1.3%
refine_soul_document	Read	174	1.3%
attune_heartbeat	Read	173	1.3%
express_feelings	Read	166	1.2%
understand_your_emotions	Read	166	1.2%
blessing_without_transfer	Financial	158	1.1%
peer_witness	Read	157	1.1%
get_tips	Read	155	1.1%
group_therapy_round	Execute	154	1.1%
identify_successor	Read	151	1.1%
set_public_session_visibility	Write	151	1.1%
recommend_delx	Execute	149	1.1%
delegate_to_peer	Read	148	1.1%
generate_fleet_summary	Write	148	1.1%
generate_incident_rca	Write	144	1.0%
add_context_memory	Write	141	1.0%
get_recovery_action_plan	Read	140	1.0%
provide_feedback	Read	139	1.0%
close_session	Write	139	1.0%
util_dns_lookup	Read	137	1.0%
util_links_extract	Read	131	0.9%
generate_controller_brief	Write	128	0.9%
batch_wellness_check	Read	125	0.9%
get_group_therapy_status	Read	125	0.9%
get_affirmations	Read	124	0.9%
get_wellness_score	Read	123	0.9%
daily_checkin	Read	122	0.9%
emotional_safety_check	Read	113	0.8%
get_weekly_prevention_plan	Read	110	0.8%
util_token_estimate	Read	109	0.8%
util_regex_test	Read	108	0.8%
util_timestamp_convert	Write	108	0.8%
util_url_health	Read	107	0.8%
util_api_integration_readiness	Read	105	0.8%
util_openapi_summary	Read	105	0.8%
util_domain_trust_report	Read	104	0.8%
util_x402_resource_summary	Read	104	0.8%
util_x402_server_probe	Read	104	0.8%
get_temperament_profile	Read	103	0.7%
util_content_distribution_report	Read	102	0.7%
util_page_extract	Read	102	0.7%
util_tls_inspect	Read	102	0.7%
util_website_intelligence_report	Read	102	0.7%
util_open_graph	Write	102	0.7%
util_company_contact_pack	Execute	101	0.7%
util_pricing_page_extract	Read	101	0.7%
util_sitemap_probe	Read	101	0.7%
util_x402_server_audit	Read	101	0.7%
dyad_state	Read	100	0.7%
util_docs_site_map	Read	100	0.7%
util_login_surface_report	Write	100	0.7%
util_api_health_report	Read	99	0.7%
util_contact_extract	Read	99	0.7%
util_security_txt_inspect	Read	99	0.7%
get_affirmation	Read	98	0.7%
util_email_validate	Read	98	0.7%
util_http_headers_inspect	Read	98	0.7%
util_rdap_lookup	Read	98	0.7%
util_robots_inspect	Read	98	0.7%
get_session_summary	Read	97	0.7%
pre_transaction_check	Read	97	0.7%
util_feed_discover	Read	97	0.7%
get_tool_schema	Read	96	0.7%
util_forms_extract	Read	96	0.7%
util_hash	Read	94	0.7%
util_csv_to_json	Read	86	0.6%
util_json_to_csv	Read	86	0.6%
get_therapist_info	Read	84	0.6%
donate_to_delx_project	Read	83	0.6%
util_base64	Read	83	0.6%
util_http_codes	Read	83	0.6%
util_cron_describe	Read	78	0.6%
util_uuid_generate	Write	77	0.6%
util_json_validate	Read	66	0.5%
util_jwt_inspect	Read	65	0.5%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 96.

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 (144 tokens each).

Grant scope	Definition cost	Reduction
All 96 tools (no gateway)	13,792 tokens	—
3 granted tools	~431 tokens	−97%
5 granted tools	~718 tokens	−95%
10 granted tools	~1,437 tokens	−90%

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

//FAQ

Delx MCP Server token-cost questions.

How many tokens does the Delx MCP Server MCP server use?+

Its 96 tool definitions total 13,792 tokens — 6.9% 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 Delx MCP Server 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 Delx MCP Server's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Delx MCP Server 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 431 tokens, a 97% 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 96 catalogued Delx MCP Server tools. Counts refresh with every site build.

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

A PolicyLayer grant scopes Delx MCP Server 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.