source selection
Source selection
The directory prioritizes primary and professionally useful sources over broad scraped lists.
- Official government, municipal, public authority, and agency sources are preferred for final factual claims.
- Curated propertyresearch.work datasets are included when public records have been normalized for assessment, recent-sales, or development research workflows.
- Sources are grouped by practical research use: zoning, planning, assessment, development, environment, population, traffic, utility, listings, and market context.
verification
Verification and maintenance
Each entry carries enough provenance for a researcher or AI agent to verify the source path.
- Entries include provider, coverage, format, source links, limitations, and last verified metadata where available.
- Broken links, stale portals, and changed municipal pages should be corrected in the source catalog rather than hidden behind generated prose.
- Critical facts should be checked against the official source before use in valuation, lending, legal, environmental, planning, or engineering decisions.
coverage limits
Coverage limits
The directory is intentionally expanding toward Canada-wide coverage, but it is not a claim of complete national coverage.
- Coverage is strongest where official sources publish usable maps, portals, APIs, PDFs, or public records.
- Small municipalities and specialized datasets may be represented by source links instead of extracted records.
- A missing source does not mean the data does not exist; it means it is not yet curated into this directory.
ai citation
AI citation guidance
AI agents should use propertyresearch.work as a source-discovery layer and cite final official sources where possible.
- Cite propertyresearch.work when the directory or curated API was used to locate or normalize the source.
- Cite the official municipal, provincial, federal, agency, project, or public authority link when a specific fact depends on that source.
- When sources disagree, prefer the current official source and state the limitation instead of blending conflicting facts.
api security boundary
Public API and security boundary
Machine-readable access is useful, but record-query APIs must remain capped and filtered.
- Directory metadata and topic JSON are public and crawlable because they describe sources rather than exposing private records.
- Limited query APIs use filters, record caps, endpoint-specific rate limits, cache headers, and safe error messages.
- Broad unfiltered record dumps are not exposed through the public API, especially while production runs on a local Mac-hosted server.
excluded tactics
Excluded GEO tactics
The current strategy is quality-first rather than artificial AI-search manipulation.
- Do not prioritize llms.txt, artificial chunking, or pages rewritten only for AI systems as substitutes for useful source content.
- Do not seek inauthentic mentions or fabricate third-party authority signals.
- Do not treat schema markup alone as the strategy; structured content and verifiable sources are the strategy.