fyi-oia-nz

name	fyi-oia-nz
description	Query FYI.org.nz OIA/LGOIMA public authorities and request records through a read-only Alaveteli API surface, including request metadata, authority lookups, request timelines, and request search.
license	MIT
compatibility	Requires Python 3.10+ and network access for live data
metadata	{"thecolab.category":"government","thecolab.source_owner":"FYI and mySociety","thecolab.source_type":"community","thecolab.auth":"none","thecolab.access_mode":"public-api","thecolab.data_class":"public","thecolab.writes":"false","thecolab.browser":"false","thecolab.risk":"low","thecolab.cache_ttl":"24h","thecolab.schema_version":"1","thecolab.skill_type":"public-api","thecolab.pack":"nz-public-data","thecolab.source_url":"https://fyi.org.nz","thecolab.allowed_domains":"fyi.org.nz","thecolab.last_verified":"2026-07-19","thecolab.health":"gated","thecolab.maintainer":"@adam91holt"}

FYI.org.nz OIA/LGOIMA

Scope

This skill wraps FYI.org.nz public authority/request read endpoints and the published all-authorities CSV. It is designed for public-corpus research and reproducible citations around OIA/LGOIMA requests.

Use this when

• You need NZ public authority discovery from FYI’s authority list
• You need authority detail for a specific FYI body URL slug
• You need request metadata for an ID or request/<id>-<slug>
• You need request event timelines (calculated_state/state transitions)
• You need simple request-text search across FYI pages as a discovery aid

Do not use this for

• Request drafting or posting
• Requester personal data extraction
• Bulk crawling of the search surface

Commands

• authorities [--tag <tag>] [--query <text>] [--limit N] [--json]

  • List public authorities from /body/all-authorities.csv, with optional tag/text filtering.

• categories [--limit N] [--json]

  • List body tags/categories discovered in all-authorities.csv.

• body <url_name_or_url> [--json]

  • Fetch authority metadata from /body/<url_name>.json.

• search <query> [--limit N] [--json]

  • Search FYI request pages for matching titles; uses /search/* endpoints where available and falls back to parsed request result HTML.

• request <id_or_slug> [--full] [--json]

  • Fetch request metadata from /request/<id>.json.
  • Metadata-only mode strips requester-identifying fields.
  • --full includes the full timeline with event/type fields and message IDs.

• events <id_or_slug> [--json]

  • Fetch and flatten request event timeline only.

Read-only contract

• No logins, tokens, keys, browser automation, or write operations.
• Defaults are metadata-first and remove requester personal text/name fields from command output.
• Uses only HTTPS GETs with conservative timeouts.

Resources

• scripts/cli.py
• scripts/smoke_test.py
• references/api-notes.md