For developers

AI Governance API

Programmatic access to Policy Window's structured AI-governance catalog — designed as an agent-readable regulatory substrate. Free public endpoints for catalog discovery, machine-readable exports, and snapshot pinning; authenticated premium scopes for change feeds, semantic diffs, and research-module reads (see /wiki/services).

OpenAPI 3.1 spec

Canonical spec at /api/openapi. Import into Postman, Insomnia, Swagger UI, or your preferred OpenAPI client. CORS-open; no auth required to read the spec itself.

Free public endpoints

All free endpoints are CORS-open, no auth required. Cached at the edge for 5 minutes.

GET /wiki/catalog/json — full structured catalog (instruments, topics, concepts, benchmarks, coverage cells, citations, confidence tiers, CC0 citation graph per I4OC).
GET /wiki/catalog/csv — one-row-per-cell expansion of the coverage matrix (RFC 4180 CSV; CSVW metadata at /wiki/catalog/csv-schema).
GET /wiki/catalog/jsonld-context — JSON-LD @context for the pw: vocabulary (RDF-compatible).
GET /wiki/[slug]/datacite.xml — per-article DataCite 4.4 metadata for Zenodo / DataCite Commons / institutional-repository ingestion.
GET /wiki/[slug]?asOf=YYYY-MM-DD — historical-snapshot pinning. See /wiki/persistent-id.
GET /wiki/changelog/feed — RSS 2.0 feed of catalog changes, ETag-cached.
GET /sitemap.xml + /robots.txt — crawler discovery.

Authenticated + commercial scopes

Premium endpoints are bearer-token-gated via scoped tokens issued from /api/auth/issue-bearer. The following scopes are documented (Batch B+ availability):

api.premium — premium exports with provision-level fields + historical snapshots
change-feed.subscribe — paginated SemanticDiff endpoint /api/change-feed
research-module.read — research-module records (ResearchModule, CaseStudy, InterviewEvidence)

Commercial licence + scope provisioning is available on inquiry — email hello@policywindow.org with your use case. See /wiki/services for the full offering description.

Stable identifiers

Instrument short-codes (e.g. EU-AIA-2024, US-EO-14110), topic codes (e.g. foundation_models), concept codes (e.g. frontier-tier), and slugs are stable under PW's naming convention. URL renames issue a 308 redirect from the old slug; the old short-code is preserved as an alias in the JSON dump. Use the short-code as your foreign key, not the URL slug.

Citation in your output

When you cite a Policy Window catalog cell in downstream output, use the ?asOf=YYYY-MM-DD URL form so the citation is permanent. The recommended form:

Policy Window. (YYYY). Article title. Retrieved YYYY-MM-DD from https://policywindow.org/wiki/[slug]?asOf=YYYY-MM-DD

Eight reference-manager export formats per article (APA, Chicago, Harvard, BibTeX, RIS, CSL-JSON, OSCOLA, Bluebook); DataCite XML at /wiki/[slug]/datacite.xml pre-stages Q4 2026 Zenodo DOI registration.

MCP / agent endpoint roadmap

An MCP (Model Context Protocol) endpoint exposing the catalog as an agent-readable substrate is on the roadmap. Until that ships, the OpenAPI 3.1 spec + the catalog JSON dump are the canonical agent-integration path — both designed to be parsable by LLM tool-use without schema-inference workarounds.

Licensing

Catalog content (article prose, coverage cells, notes, citations): CC BY 4.0. Source code (this Next.js application): MIT. Citation graph specifically (structured citation triples): CC0 1.0 per the I4OC norm. See charter §5.

Important — what Policy Window is and isn't

Not legal advice. Policy Window outputs are evidence references, not legal counsel. For jurisdictional interpretation, regulatory-strategy guidance, or compliance opinions, consult qualified counsel.
Not court-citation-grade authority. Catalog rows + coverage cells point to primary sources at the provision level where available; verify against the authoritative regulator publication before quoting in a filing.
Not real-timeunless a specific surface is labelled as such. Catalog rows refresh on an editorial cadence; freshness is signalled per-article via the "Last verified" chip and the ?asOf=YYYY-MM-DD permanent-citation parameter.
AI-generated analysis is labelled. Any output produced by the AGI Social Scientist research engine carries an explicit AI-output badge; final editorial approval is always a human decision.
Human-validated outputs are labelled separately. Editorial-board approval governs catalog content; the on-demand expert validation bench reviews research-module outputs. The two layers are distinct and the validation badge on each output names which (if either) has signed off.

These disclaimers apply across every public Policy Window surface. The full operating commitments are at /wiki/charter; the AI-use map is at /wiki/ai-disclosure.

AI Governance API

Free public endpoints

All free endpoints are CORS-open, no auth required. Cached at the edge for 5 minutes.

GET /wiki/catalog/json — full structured catalog (instruments, topics, concepts, benchmarks, coverage cells, citations, confidence tiers, CC0 citation graph per I4OC).

GET /wiki/catalog/csv — one-row-per-cell expansion of the coverage matrix (RFC 4180 CSV; CSVW metadata at /wiki/catalog/csv-schema).

GET /wiki/catalog/jsonld-context — JSON-LD @context for the pw: vocabulary (RDF-compatible).

GET /wiki/[slug]/datacite.xml — per-article DataCite 4.4 metadata for Zenodo / DataCite Commons / institutional-repository ingestion.

GET /wiki/[slug]?asOf=YYYY-MM-DD — historical-snapshot pinning. See /wiki/persistent-id.

GET /wiki/changelog/feed — RSS 2.0 feed of catalog changes, ETag-cached.

GET /sitemap.xml + /robots.txt — crawler discovery.