Skip to main content
Glama

Queue a building permit CSV export

create_permit_export

Queues a CSV of building permits from exactly one source: a saved permit alert (alertId, palr_…; alertScope picks not_dismissed (the default), pending, saved (kept), dismissed, or all), a list (listId), permitIds already in hand (at most 10,000), or a new filter set (at most 10,000 permits). Filters cannot be combined with an alert, list, or ids; to narrow one, pull its permits with the narrowing filters first and export those permitIds. Every row has Latitude, Longitude, and Location Precision columns for loading into a map. deliveryMode="new" leaves out permits already delivered, including ones already read in this chat. The response is a receipt with exportJobId, never the CSV bytes. Queuing does not use credits; new or changed rows count toward 100,000 export records a month on County, State, and National, and cost 1 credit each on API Only and the trial. Rows already delivered unchanged are not charged again. Permit contacts are in the file. count_permits returns the match count for a new filter set. get_export returns status, and the file downloads from https://boardwalkai.com/exports.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "create_permit_export"
  }
}

Input Schema

NameRequiredDescriptionDefault
qNoSame as the permits map search box: words from the permit (work description, address, names) or a permit number. One search text, not a list: for several alternative terms, make one call per term. Use one of q, permitNumber, or contractor per call.
idsNoAlias of permitIds.
bboxNoMap area as "west,south,east,north" or {west,south,east,north}. Around a planned project, search q with its street address first; use a small box around its coordinates only when it has no address, and say which permits are at neighbouring addresses.
citiesNoRegion: city ids or names ("Lehi"). A permit in any listed state, county, or city matches, so do not add the county. A name that matches several cities is refused with each candidate's id and state: pass the id, or add the state (states:["UT"]), which then only helps find the city.
dateToNoWindow end, YYYY-MM-DD, inclusive.
listIdNoExport the permits saved on this list ("lst_…"). Not with filters.
stagesNoStatus stage as "<status>:<stage>" or "bucket:<status>"; get_permit_reference facet "stages" lists them.
statesNoRegion: state ids, names, or abbreviations ("UT"). states, counties, and cities are one Region filter: a permit in any of them matches. Send only the narrowest place, never "UT" together with "Utah County".
tradesNoTrades the job involves (the scope of the work, not open bids): site_work, foundation, paving_concrete, framing, roofing, windows_doors, siding_exterior, insulation, electrical, plumbing, hvac, gas, fire_protection, low_voltage, solar, elevator, drywall, flooring, cabinets_countertops, finish_carpentry, painting, appliances, cleaning, landscaping, fencing, pool. roofing here also matches new homes and additions with roof work; workTypes roofing is re-roof jobs only. When a trade rep asks for "roofing permits", count both and say which is which.
whyNowNoWhy now codes: issued_no_inspections, rough_in_passed, finals_next, last_inspection_failed, on_hold, fees_due, slow_review, new_owner, expiring_soon.
alertIdNoExport the permits on this permit alert ("palr_…"). Not with filters.
sectorsNoPrivate or public work. Default both.
sqftMaxNoSquare feet maximum.
sqftMinNoSquare feet minimum.
countiesNoRegion: county ids or names ("Utah County"). Names that match more than one place are refused, not guessed; the refusal lists each candidate's id and state. A permit in any listed state, county, or city matches. A state sent with a county inside it only helps find the county (warning region_parent_state_dropped).
dateFromNoWindow start, YYYY-MM-DD, inclusive.
filenameNoOptional file name, 1–120 characters.
jobKindsNoMajor or minor jobs. Default both; "major" alone leaves out small jobs such as a water heater swap.
placeKeyNoplaceKey from a permit: every permit at that address (at most 20 keys).
statusesNoStatus: in_review (applied or approved, not issued yet), active (issued, under way, or on hold), final (completed or closed), inactive (ended without a final: expired, withdrawn, denied, void, or stopped), no_construction (a record with no construction work).
unitsMaxNoUnits maximum.
unitsMinNoUnits minimum (homes or apartments on the permit).
valueMaxNoJob value maximum in dollars.
valueMinNoJob value minimum in dollars, as stated on the permit. "Over $50k" → 50000.
dateFieldNoWhich date the window applies to: applied, issued, or activity (latest activity, the default).
dateRangeNoDate window. Defaults to 12m. "all" removes the window.
permitIdsNoPermit ids already in hand, at most 10,000.
workTypesNoWork type keys: new_home, multifamily, adu, manufactured, new_commercial, new_other, addition, interior_remodel, kitchen_bath, tenant_improvement, accessory, structural_repair, repair, hvac, electrical, solar_battery, generator, plumbing, water_heater, gas, fire_protection, telecom, roofing, windows_doors, siding, pool, fence, signage, demolition, grading, utilities, driveway_sidewalk, septic_well, planning, unspecified. A work type is what the job is: roofing = a re-roof or roof repair job. For every job that includes a trade (a new home includes roofing), use trades.
alertScopeNoWith alertId: not_dismissed (the default), pending, saved (kept), dismissed, or all.
contractorNoA contractor's full name, e.g. "Acme Roofing". Case, punctuation, and endings such as LLC, Inc., Corp., Co., and Ltd. are ignored, so "acme roofing" also finds "Acme Roofing, LLC". Every other word counts, so part of a name ("Acme") matches nothing. When unsure of the name, call search_permit_contractors (q with the name words) and pass its contractorKeys.
detailOnlyNoOnly permits whose full detail page was read.
maxCreditsNoOptional ceiling on credits this call may spend (API Only and the trial). The API refuses rather than exceed it.
valueBandsNoValue bands: u10k (under $10k), 10k, 50k, 250k, 1m (and up).
workGroupsNoWork type groups: new (new homes and buildings), remodel, systems, exterior, site, planning, unspecified.
assetClassesNoBuilding type class (3rd level), e.g. Single-Family Home.
deliveryModeNo"new" leaves out permits already delivered to this account. Default "all".
permitNumberNoAn exact permit number, e.g. "BLD-2025-01234". Searches every date unless you send a date filter.
hasAnyContactNoOnly permits with any phone or email, private individuals included.
hasContractorNoOnly permits that name a contractor.
propertyTypesNoBuilding type, the same labels as planned projects (get_project_types): Residential, Commercial, …
contractorKeysNoExact contractor keys from search_permit_contractors (at most 50). Every permit by those contractors.
hasInspectionsNoOnly permits with at least one inspection that has a result (passed, failed, or completed).
includeRecordsNoAlso return non-job records (paperwork on an existing building). Default false.
propertySubtypesNoMore-specific building type (2nd level).
constructionTypesNoConstruction type: new_building, addition, renovation, site_work, infrastructure, demolition_only, none, not_stated.
hasCompanyContactNoOnly permits with a company phone or email printed on them.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.8/5.0
Behavior5/5

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

Annotations only say this is a non-read-only, non-destructive write, but the description adds far more: the response is a receipt with exportJobId and never the CSV bytes, where the file downloads, that queuing is free while new/changed rows are metered (100,000/month on County/State/National, 1 credit each on API Only and trial), and that unchanged re-delivered rows are not re-charged. That is exactly the cost and return-surface context an agent needs before committing.

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?

The source enumeration is front-loaded and every sentence carries operational weight (row columns, receipt-only response, cost model, download URL). The single dense paragraph with heavy parenthetical nesting makes the cost rules and region guidance harder to scan than they need to be, costing one point.

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

Completeness5/5

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

For a 46-parameter tool with no output schema, the description covers the full lifecycle: what is queued, that the return is a receipt rather than data, the guaranteed Latitude/Longitude/Location Precision columns, contact inclusion, billing, and the follow-up call and download location. Nothing needed to call it correctly or interpret the result is missing.

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 an extra point by adding cross-parameter semantics the schema cannot express, notably the 10,000 cap on permitIds/ids and filters, alertScope behavior tied to alertId, and deliveryMode="new" excluding rows already read in this chat. It does not restate every field, which is appropriate given the schema already documents all 46.

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?

Opens with a specific verb and resource ("Queues a CSV of building permits") and immediately constrains scope to exactly one of four sources, naming each by parameter. An agent can distinguish this from create_export, count_permits, and get_export without opening another 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?

Explicitly states the mutual-exclusion rule ("Filters cannot be combined with an alert, list, or ids") and prescribes the workaround: pull permits with the narrowing filters, then export those permitIds. It also routes counting to count_permits and status/download to get_export, so the when-to-use picture is complete.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources