nzx
Query NZX public delayed market data through a lightweight no-login CLI. Use when the task involves New Zealand Exchange listed share prices, S&P/NZX index levels, NZSX movers, historical daily OHLC-s
SKILL.mdskills/nzx/SKILL.md
GitHubNZX
Goal
Query NZX public delayed market data through a small deterministic CLI with human-readable and JSON output, without browser automation, login, API keys, portfolio access, or local cache state.
This skill ships support for:
- Delayed quote snapshots for one or more NZX tickers
- S&P/NZX 50, S&P/NZX 20, and S&P/NZX All index snapshots
- NZSX top gainers and losers
- Two-year daily OHLC-style instrument performance with volume/activity fields
- Upcoming and recent dividends
- NZSX ticker/company search
Use this when
- A user asks for current delayed NZX-listed share or ETF prices
- A user asks for NZX 50, NZX 20, or NZX All index levels
- A user wants today's NZSX gainers or losers
- A user wants daily price history, high/low/open/close, volume, or market cap for an NZX ticker
- A user asks about recent or upcoming NZX dividends
- A user wants to find NZX tickers by company or instrument name
- A workflow needs public, no-login, machine-readable NZX market data
Do not use this for
- Trading, order placement, portfolio holdings, broker login, watchlists, or account actions
- Real-time market-data redistribution or claims that values are live tick-by-tick
- Investment advice, valuation recommendations, tax advice, or suitability decisions
- Non-NZX exchanges unless the instrument is surfaced by NZX public data
- Paid NZX data products or authenticated vendor feeds
Preferred workflow
- Run
scripts/cli.pywith the narrowest subcommand that answers the task - Use
quotefor delayed ticker snapshots andindexfor S&P/NZX index levels - Use
movers --topormovers --losersfor market-wide daily movers - Use
history <ticker> --from ... --to ...for daily OHLC-style history - Use
dividends <ticker>for upcoming and past dividend rows - Use
search <query>to resolve company names to ticker codes before quote/history calls - Use
--jsonwhen another tool or agent needs machine-readable output - Mention that NZX market data is delayed by 20 minutes and may be cached
CLI
Run with:
python3 skills/nzx/scripts/cli.py <command> [flags]
Commands
quote <ticker> [<ticker>...] [--json]- delayed price snapshot for one or more tickersindex [--name nzx50|nzx20|nzx-all] [--json]- delayed index level and day changemovers [--top|--losers] [--limit N] [--json]- top NZSX gainers or losers todayhistory <ticker> [--from DATE] [--to DATE] [--limit N] [--json]- daily OHLC-style performance historydividends <ticker> [--json]- upcoming and recent dividendssearch <query> [--limit N] [--json]- find NZSX tickers by code or company/instrument name
Examples:
python3 skills/nzx/scripts/cli.py quote AIR FPH MEL --json
python3 skills/nzx/scripts/cli.py index --name nzx50 --json
python3 skills/nzx/scripts/cli.py movers --losers --limit 5
python3 skills/nzx/scripts/cli.py history AIR --from 2026-05-01 --to 2026-05-22 --json
python3 skills/nzx/scripts/cli.py dividends AIR --json
python3 skills/nzx/scripts/cli.py search "a2 milk" --json
Resources
- CLI entrypoint:
scripts/cli.py - API and stability notes:
references/api-notes.md
Notes
- No API key, username, password, account cookie, broker login, or browser session is required for implemented read-only operations
- Upstream source is NZX:
https://www.nzx.com,https://api.nzx.com/ticker/1.0, andhttps://api.nzx.com/public - NZX states that market data is delayed by 20 minutes and page information may be cached; include that caveat in user-facing market-data answers
quote,index,movers, andsearchuse delayed ticker snapshot endpoints underhttps://api.nzx.com/ticker/1.0historyanddividendsuse public NZX market-data endpoints underhttps://api.nzx.com/public- The live SSE-style ticker endpoints are intentionally not used by the CLI because they stream indefinitely; the CLI uses
/lastsnapshot endpoints so commands terminate - Default output is human-readable; every data command supports
--jsonfor chaining into other tools - Avoid high-volume polling; use narrow ticker lists, small mover limits, and bounded history ranges