Windsock

Search the aircraft registry

search_aircraft

Read-only

Registry-wide aircraft search — the fleet / market-research search surface. Returns individual registered tails matching an AND-combined filter set (make/model, US state, city, year, tail number, current registrant name, or geographic radius), with pagination and sorting. Resolve each model name to a make_model_id with search_make_models first; to cover several models in one search pass them as a list with condition "in". For population totals (e.g. "how many R182s in Oregon") prefer count_aircraft, which avoids paging. Each filter is {field_name, condition, value}.

Input Schema

TableJSON Schema

Name	Required	Description
`limit`	No	Rows per page, 1–200 (default 25).
`offset`	No	Pagination offset (default 0).
`filters`	No	AND-combined filters; each is {field_name, condition, value}. Fields: make_model_id (numeric id), state (2-letter US code), city, year, tail_number, registrant_name (current FAA registrant / registered owner; substring via contains), location_point, registration_authority, registration_country, airworthiness_class. Conditions: is, is_not, contains, does_not_contain, in, not_in, lt, lte, gt, gte. For location_point use condition "within_<miles>" (e.g. "within_50") with value "lat,lon" (e.g. "45.52,-122.68"). To search SEVERAL make/models at once, use one clause with condition "in" and an array of ids (e.g. {field_name: "make_model_id", condition: "in", value: [1699, 1704]}) — that is a union. Do NOT repeat make_model_id in separate clauses: filters are AND-combined, so two of them match nothing. Spec filters (gt/gte/lt/lte only): engine_power_hp, useful_load_lbs (pounds); engine_make is text (is/contains, e.g. "LYCOMING"). ⚠️ These come from per-model performance specs and are NOT populated for every aircraft: engine power 75% of the fleet, weights 76%, engine make 78%. Filtering on one EXCLUDES every aircraft with no value, so tell the user that when you use them. airworthiness_class (is/is_not/in/not_in) is the FAA classification from the aircraft's LATEST registration — this is how you search for EXPERIMENTAL / amateur-built aircraft. Tokens: standard, limited, restricted, experimental, provisional, multiple, primary, special_flight_permit, light_sport. ⚠️ 36% of the fleet has no certification recorded, so filtering on it excludes all of those; say so when you use it. 🔴 STATE OF REGISTRY: this registry is NOT US-only. ~115k aircraft (21.6% of it) are on one of 17 international registers, and they are concentrated by model — 41% of the R44 fleet, 21% of C152s, 10% of SR22s. An unfiltered count is a WORLD count, not a US one. Use registration_authority with an authority code (FAA, TCCA, CASA, ANAC, DGAC, AESA, DGACL, ILT, ACG, IAA, CAANO, NSAT, CAAI, CAABG, CAALV, DAC, CAAS, ECAA) or the cohort tokens "domestic" / "international"; or registration_country with an ISO-2 code or country name ("CA", "Canada"). Say which fleet a number covers whenever you report one. ⚠️ There is NO `country` filter. The v3 validator rejects one (`country` is not a CAVN column), and the underlying value is the FAA registrant's mailing address, NULL for every international aircraft — so it could not select them even if it were accepted. Use registration_country. ⚠️ International aircraft have registry identity but carry no predicted_price and cannot be valued by tail.
`sort_field`	No	Sort key (default last_airborne_at).
`sort_direction`	No	Default desc.

Output Schema

TableJSON Schema

Name	Required	Description	Default
`data`	Yes	Matching registered tails.
`meta`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed3 schema fields changedOct 20, 2026

changedInput schema / properties / filters / description

Previous value: -"AND-combined filters; each is {field_name, condition, value}. Fields: make_model_id (numeric id), state (2-letter US code), city, year, tail_number, registrant_name (current FAA registrant / registered owner; substring via contains), seen_on_market (\"true\"/\"false\"), latest_list_date, location_point, registration_authority, registration_country, airworthiness_class. Conditions: is, is_not, contains, does_not_contain, in, not_in, lt, lte, gt, gte. For location_point use condition \"within_<miles>\" (e.g. \"within_50\") with value \"lat,lon\" (e.g. \"45.52,-122.68\"). To search SEVERAL make/models at once, use one clause with condition \"in\" and an array of ids (e.g. {field_name: \"make_model_id\", condition: \"in\", value: [1699, 1704]}) — that is a union. Do NOT repeat make_model_id in separate clauses: filters are AND-combined, so two of them match nothing. Spec filters (gt/gte/lt/lte only): engine_power_hp, useful_load_lbs (pounds), smoh_hours (engine hours since major overhaul); engine_make is text (is/contains, e.g. \"LYCOMING\"). ⚠️ These come from per-model performance specs and listing data and are NOT populated for every aircraft: engine power 75% of the fleet, weights 76%, engine make 78%, smoh_hours only 12% (84% of aircraft that have ever been listed — pair it with seen_on_market). Filtering on one EXCLUDES every aircraft with no value, so tell the user that when you use them. airworthiness_class (is/is_not/in/not_in) is the FAA classification from the aircraft's LATEST registration — this is how you search for EXPERIMENTAL / amateur-built aircraft. Tokens: standard, limited, restricted, experimental, provisional, multiple, primary, special_flight_permit, light_sport. ⚠️ 36% of the fleet has no certification recorded, so filtering on it excludes all of those; say so when you use it. 🔴 STATE OF REGISTRY: this registry is NOT US-only. ~115k aircraft (21.6% of it) are on one of 17 international registers, and they are concentrated by model — 41% of the R44 fleet, 21% of C152s, 10% of SR22s. An unfiltered count is a WORLD count, not a US one. Use registration_authority with an authority code (FAA, TCCA, CASA, ANAC, DGAC, AESA, DGACL, ILT, ACG, IAA, CAANO, NSAT, CAAI, CAABG, CAALV, DAC, CAAS, ECAA) or the cohort tokens \"domestic\" / \"international\"; or registration_country with an ISO-2 code or country name (\"CA\", \"Canada\"). Say which fleet a number covers whenever you report one. ⚠️ There is NO `country` filter. The v3 validator rejects one (`country` is not a CAVN column), and the underlying value is the FAA registrant's mailing address, NULL for every international aircraft — so it could not select them even if it were accepted. Use registration_country. ⚠️ International aircraft have registry identity but no US market history, so they carry no predicted_price and cannot be valued by tail."New value: +"AND-combined filters; each is {field_name, condition, value}. Fields: make_model_id (numeric id), state (2-letter US code), city, year, tail_number, registrant_name (current FAA registrant / registered owner; substring via contains), location_point, registration_authority, registration_country, airworthiness_class. Conditions: is, is_not, contains, does_not_contain, in, not_in, lt, lte, gt, gte. For location_point use condition \"within_<miles>\" (e.g. \"within_50\") with value \"lat,lon\" (e.g. \"45.52,-122.68\"). To search SEVERAL make/models at once, use one clause with condition \"in\" and an array of ids (e.g. {field_name: \"make_model_id\", condition: \"in\", value: [1699, 1704]}) — that is a union. Do NOT repeat make_model_id in separate clauses: filters are AND-combined, so two of them match nothing. Spec filters (gt/gte/lt/lte only): engine_power_hp, useful_load_lbs (pounds); engine_make is text (is/contains, e.g. \"LYCOMING\"). ⚠️ These come from per-model performance specs and are NOT populated for every aircraft: engine power 75% of the fleet, weights 76%, engine make 78%. Filtering on one EXCLUDES every aircraft with no value, so tell the user that when you use them. airworthiness_class (is/is_not/in/not_in) is the FAA classification from the aircraft's LATEST registration — this is how you search for EXPERIMENTAL / amateur-built aircraft. Tokens: standard, limited, restricted, experimental, provisional, multiple, primary, special_flight_permit, light_sport. ⚠️ 36% of the fleet has no certification recorded, so filtering on it excludes all of those; say so when you use it. 🔴 STATE OF REGISTRY: this registry is NOT US-only. ~115k aircraft (21.6% of it) are on one of 17 international registers, and they are concentrated by model — 41% of the R44 fleet, 21% of C152s, 10% of SR22s. An unfiltered count is a WORLD count, not a US one. Use registration_authority with an authority code (FAA, TCCA, CASA, ANAC, DGAC, AESA, DGACL, ILT, ACG, IAA, CAANO, NSAT, CAAI, CAABG, CAALV, DAC, CAAS, ECAA) or the cohort tokens \"domestic\" / \"international\"; or registration_country with an ISO-2 code or country name (\"CA\", \"Canada\"). Say which fleet a number covers whenever you report one. ⚠️ There is NO `country` filter. The v3 validator rejects one (`country` is not a CAVN column), and the underlying value is the FAA registrant's mailing address, NULL for every international aircraft — so it could not select them even if it were accepted. Use registration_country. ⚠️ International aircraft have registry identity but carry no predicted_price and cannot be valued by tail."

changedInput schema / properties / filters / items / properties / field_name / enum

Previous value: -[
-  "make_model_id",
-  "state",
-  "city",
-  "year",
-  "tail_number",
-  "registrant_name",
-  "seen_on_market",
-  "latest_list_date",
-  "location_point",
-  "engine_power_hp",
-  "useful_load_lbs",
-  "engine_make",
-  "smoh_hours",
-  "airworthiness_class",
-  "registration_authority",
-  "registration_country"
-]New value: +[
+  "make_model_id",
+  "state",
+  "city",
+  "year",
+  "tail_number",
+  "registrant_name",
+  "location_point",
+  "engine_power_hp",
+  "useful_load_lbs",
+  "engine_make",
+  "airworthiness_class",
+  "registration_authority",
+  "registration_country"
+]

changedInput schema / properties / sort_field / enum

Previous value: -[
-  "last_airborne_at",
-  "year",
-  "tail_number",
-  "last_action_date",
-  "latest_list_date"
-]New value: +[
+  "last_airborne_at",
+  "year",
+  "tail_number",
+  "last_action_date"
+]

First observedSep 11, 2026

TDQS

A4.3/5.0

Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=true and destructiveHint=false, so safety is covered; the description adds return shape (individual tails), pagination/sorting, and the union-vs-intersection trap. However, the genuinely consequential behavioral disclosures — spec filters excluding 22-25% of the fleet, 36% missing airworthiness_class, 21.6% international aircraft meaning an unfiltered count is a world count, and international tails having no predicted_price — live in the filters schema text rather than the tool description, so the description itself carries only a moderate share of the burden.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with purpose, then routing, then a note on filter shape — a sensible order. The closing 'Each filter is {field_name, condition, value}' mildly restates the schema and is the only near-redundant line in an otherwise dense, waste-free paragraph.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a complex filter-search surface with an output schema, this covers purpose, return granularity, alternatives, and union semantics adequately. The one residual gap is that the most costly misuse — reporting a worldwide count as a US total, or using a nonexistent `country` filter — is only guarded inside the filters schema description, not at the top level where an agent skims first.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3; the description earns above that by explaining cross-parameter semantics the schema only implies — that filters are AND-combined, that multi-model search must be a single 'in' clause with an id array, and that each filter is {field_name, condition, value}. It does not itself mention limit/offset ceilings or sort defaults, leaving those to the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Registry-wide aircraft search') and immediately scopes the output ('Returns individual registered tails'), which distinguishes it from the aggregate sibling count_aircraft. It also names the filter axes and the surrounding tool chain (search_make_models for id resolution), so an agent can place it in the workflow without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit routing rules: resolve model names via search_make_models first, use condition 'in' to union several models in one pass, and prefer count_aircraft for population totals like 'how many R182s in Oregon'. It even pre-empts the failure mode of repeating make_model_id in separate AND-combined clauses.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Windsock tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.