/api/v1/pipeline/search?q=yonge&status=builtSearch applications by address, id, or neighbourhood
{
"count": 40,
"results": [
{ "id": "17 122573 STE 20 OZ",
"address": "…",
"status": "built",
"units": 7190 }
]
}An open-source civic project by Nshipyard. Not affiliated with the Government of Canada or the City of Toronto.
Nshipyard Canada · Project 08
City Planning publishes a Development Pipeline: every large project from application to occupancy, with proposed unit counts. This is that file cleaned, joined to wards and neighbourhoods, with a normalized status taxonomy, plus building-permit issuance times from 202,779 active permits.
2,391
development applications in the official pipeline, each with a normalized status
124,326
homes in projects marked Built: the net homes actually gained
803,206
homes still in the pipeline: proposed or active, not yet built
272
median days for a New Building permit to go from application to issuance
Explorer
Search above, or filter by status or ward, to see an application's units, timeline, and neighbourhood.
Showcase
The pipeline file lists proposed units for every project. Counting only projects the City marks Built gives the net homes actually gained, by 158-model neighbourhood. Everything else is still pipeline: proposed or active, not yet built.
Darker red means more Built homes. 96.2% of pipeline records were matched to a neighbourhood.
Showcase
From 202,779 active building permits, the calendar days between application and issuance, by permit type. A New Building permit takes a median of 272 days. A small residential project takes 19. The gap is the story.
This dataset is a snapshot of permits active as of October 2026. Older years only include permits still open, so early-year medians carry survivorship bias.
For developers
Three consumption paths, same normalized data. REST for applications, OpenAPI for integration, MCP tools over streamable HTTP for AI agents.
/api/v1/pipeline/search?q=yonge&status=builtSearch applications by address, id, or neighbourhood
{
"count": 40,
"results": [
{ "id": "17 122573 STE 20 OZ",
"address": "…",
"status": "built",
"units": 7190 }
]
}/api/v1/pipeline/permit_times?type=New%20BuildingMedian issuance days by permit type and year
{
"count": 23,
"results": [
{ "permit_type": "New Building",
"year": 2024, "n": 149,
"median_days": 184.0 }
]
}/api/v1/pipeline/summaryTotals, units by status, built units by neighbourhood
{
"records": 2391,
"built_units": 124326,
"units_by_status": {
"proposed": 367469,
"active": 435737,
"built": 124326 }
}Connect your agent
Pick your harness, copy the prompt, send it to your agent. Your agent runs the setup itself.
Set up the Toronto Development Pipeline MCP server so I can query it from here. 1. Run: claude mcp add --transport http toronto-pipeline https://this-site.example/mcp 2. Run `claude mcp list` to confirm it connected. 3. Look up the development application with the most built homes and tell me its address and unit count, and show me the result.
Methodology
Source: the City of Toronto's official Development Pipeline analytical dataset, retrieved October 8, 2026, joined to the Development Applications file on application number (96.8% match) for coordinates, then to the 158-model neighbourhood boundaries (96.2% matched).
Status taxonomy: Under Review becomes proposed (submitted, decision pending); Active becomes active (approved or under construction, not yet recorded complete); Built becomes built (completed). Only built units count as homes gained.
Unit counts are proposed units from planning applications, not final occupancy counts. The pipeline covers larger developments requiring Planning Act approvals; as-of-right development below the Site Plan Control threshold is excluded.
Permit times come from the Building Permits Active Permits file: calendar days from application date to issued date. It is a snapshot of permits active in October 2026, so older years only include permits still open. Negative values and values over 10 years were excluded as data-entry errors.
The City does not publish per-project completion dates or final built unit counts in these files; built status is the closest available proxy for homes gained.
Data
Versioned releases, MIT licensed. CSV for spreadsheets.