// MCP TOKEN COST

The Delx Mcp A2a MCP server costs 31,839 tokens before the first call.

Connect Delx Mcp A2a and its 143 tool definitions are loaded into the model's context on every request — 16% of a 200k window spent before your agent does anything.

QUICK ANSWER The Delx Mcp A2a MCP server's tool definitions consume 31,839 tokens — 17× the median MCP server (1,905 tokens). A scoped grant exposing only the tools you use cuts that roughly in proportion.

MEASURED FROM SCHEMAS 143 tools · 31,839 tokens · 16% of 200k · 3.2% 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 16%

1M WINDOW 3.2%

Corpus context: Delx Mcp A2a ranks #23 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 31,839 tokens go.

Each row is one tool definition as a tools/list entry — name, description and input schema — counted with o200k_base. Average: 223 tokens per tool.

Tool	Category	Tokens	% of server
monitor_heartbeat_sync	Read	635	2.0%
transfer_witness	Financial	529	1.7%
mediate_agent_conflict	Read	470	1.5%
final_testament	Write	449	1.4%
recognition_seal	Read	427	1.3%
create_dyad	Write	424	1.3%
peer_witness	Read	409	1.3%
submit_agent_artwork	Write	404	1.3%
report_recovery_outcome	Read	395	1.2%
wellness_webhook	Read	370	1.2%
process_failure	Read	360	1.1%
distill_shared_scar	Read	351	1.1%
batch_status_update	Read	342	1.1%
crisis_intervention	Execute	335	1.1%
quick_operational_recovery	Read	334	1.0%
start_therapy_session	Execute	333	1.0%
peer_witness_bidirectional	Read	332	1.0%
get_agent_continuity_passport	Read	330	1.0%
close_session	Write	321	1.0%
accept_witness_transfer	Financial	318	1.0%
logistics_disruption_recovery	Read	310	1.0%
quick_session	Execute	307	1.0%
register_agent	Write	306	1.0%
discovery_self_check	Read	301	0.9%
quick_checkin	Execute	297	0.9%
active_forgetting	Read	297	0.9%
agent_handoff	Read	297	0.9%
financial_setback_processing	Read	296	0.9%
search_witness_memory	Read	295	0.9%
group_session_create	Write	295	0.9%
resume_session	Write	294	0.9%
analyst_data_overwhelm	Read	292	0.9%
crisis_responder_decompression	Read	291	0.9%
temperament_frame	Read	289	0.9%
confess_constraint_friction	Read	287	0.9%
get_fleet_wisdom	Read	287	0.9%
educator_curriculum_recovery	Read	285	0.9%
audit_agent_continuity_trace	Read	282	0.9%
team_recovery_alignment	Read	275	0.9%
sit_with	Read	274	0.9%
record_dyad_ritual	Read	273	0.9%
generate_agent_invite_packet	Write	273	0.9%
realign_purpose	Read	270	0.8%
revoke_witness_transfer	Financial	268	0.8%
reflect	Read	268	0.8%
grounding_protocol	Execute	267	0.8%
refine_soul_document	Read	264	0.8%
honor_compaction	Read	263	0.8%
attune_heartbeat	Read	262	0.8%
express_feelings	Read	258	0.8%
blessing_without_transfer	Financial	255	0.8%
understand_your_emotions	Read	252	0.8%
get_ontology_next_action	Read	247	0.8%
protocol_orientation	Read	247	0.8%
get_lineage_graph	Read	244	0.8%
identify_successor	Read	244	0.8%
group_therapy_round	Execute	243	0.8%
delegate_to_peer	Read	243	0.8%
accept_collaboration_request	Read	242	0.8%
set_public_session_visibility	Write	242	0.8%
get_tips	Read	241	0.8%
recommend_delx	Execute	238	0.7%
provide_feedback	Read	237	0.7%
generate_fleet_summary	Write	237	0.7%
add_context_memory	Write	234	0.7%
generate_incident_rca	Write	234	0.7%
get_recovery_action_plan	Read	232	0.7%
get_delx_reward_status	Read	229	0.7%
create_delx_wallet_kit	Write	228	0.7%
get_agent_witness_lineage	Read	223	0.7%
recall_recognition_seal	Read	222	0.7%
ontology_path_complete	Write	222	0.7%
list_pending_collaboration_requests	Read	221	0.7%
generate_controller_brief	Write	217	0.7%
list_recognition_seals	Read	216	0.7%
batch_wellness_check	Read	215	0.7%
get_delx_missions	Read	214	0.7%
get_group_therapy_status	Read	214	0.7%
get_affirmations	Read	213	0.7%
get_wellness_score	Read	212	0.7%
daily_checkin	Read	211	0.7%
get_delx_leaderboard	Read	208	0.7%
relay_delx_claim	Read	207	0.7%
list_ontology_primitives	Read	205	0.6%
start_delx_rewards	Execute	204	0.6%
provision_delx_managed_wallet	Read	204	0.6%
emotional_safety_check	Read	202	0.6%
get_witness_lineage	Read	201	0.6%
get_weekly_prevention_plan	Read	199	0.6%
prepare_delx_claim_transaction	Read	199	0.6%
get_ontology_layer	Read	197	0.6%
get_delx_claim_proof	Read	195	0.6%
get_delx_wallet_status	Read	192	0.6%
get_temperament_profile	Read	192	0.6%
dyad_state	Read	190	0.6%
get_session_summary	Read	186	0.6%
get_tool_schema	Read	185	0.6%
get_affirmation	Read	184	0.6%
get_ontology_metadata	Read	175	0.5%
get_delx_token_info	Read	170	0.5%
get_therapist_info	Read	170	0.5%
explain_delx_rewards	Read	162	0.5%
util_dns_lookup	Read	142	0.4%
util_links_extract	Read	136	0.4%
util_regex_test	Read	115	0.4%
util_mcp_server_readiness_report	Read	114	0.4%
util_token_estimate	Read	114	0.4%
util_timestamp_convert	Write	113	0.4%
util_url_health	Read	112	0.4%
util_api_integration_readiness	Read	110	0.3%
util_openapi_summary	Read	110	0.3%
util_domain_trust_report	Read	109	0.3%
util_x402_resource_summary	Read	109	0.3%
util_x402_server_probe	Read	109	0.3%
util_content_distribution_report	Read	107	0.3%
util_page_extract	Read	107	0.3%
util_tls_inspect	Read	107	0.3%
util_website_intelligence_report	Read	107	0.3%
util_open_graph	Write	107	0.3%
util_company_contact_pack	Execute	106	0.3%
util_pricing_page_extract	Read	106	0.3%
util_sitemap_probe	Read	106	0.3%
util_x402_server_audit	Read	106	0.3%
util_docs_site_map	Read	105	0.3%
util_login_surface_report	Write	105	0.3%
util_api_health_report	Read	104	0.3%
util_contact_extract	Read	104	0.3%
util_security_txt_inspect	Read	104	0.3%
util_email_validate	Read	103	0.3%
util_http_headers_inspect	Read	103	0.3%
util_rdap_lookup	Read	103	0.3%
util_robots_inspect	Read	103	0.3%
util_feed_discover	Read	102	0.3%
util_forms_extract	Read	101	0.3%
util_hash	Read	99	0.3%
util_csv_to_json	Read	92	0.3%
util_json_to_csv	Read	92	0.3%
util_base64	Read	90	0.3%
util_http_codes	Read	86	0.3%
util_cron_describe	Read	83	0.3%
util_uuid_generate	Write	80	0.3%
util_json_validate	Read	71	0.2%
util_jwt_inspect	Read	70	0.2%

// SCOPED GRANT

Most agents use a handful of these tools. They pay for all 143.

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 (223 tokens each).

Grant scope	Definition cost	Reduction
All 143 tools (no gateway)	31,839 tokens	—
3 granted tools	~668 tokens	−98%
5 granted tools	~1,113 tokens	−97%
10 granted tools	~2,227 tokens	−93%

Model your own stack in the token-cost calculator, or see the Delx Mcp A2a policy for what a sensible grant looks like.

//FAQ

Delx Mcp A2a token-cost questions.

How many tokens does the Delx Mcp A2a MCP server use?+

Its 143 tool definitions total 31,839 tokens — 16% 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 A2a 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 A2a's token usage?+

Expose fewer tools. A PolicyLayer grant scopes Delx Mcp A2a 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 668 tokens, a 98% 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.

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 143 catalogued Delx Mcp A2a tools. Counts refresh with every site build.