List PoE2 leagues available on the trade API, and which are collected.

Use this first when the user names a league loosely ("the new league"), so later calls pass an exact league id.

Report collector health, database coverage and remaining rate budget.

Check this before concluding that missing data means a missing market — an empty history usually means the collector has not been running.

Find tracked items by name, returning the item keys other tools take.

Every priceable thing has a stable key (cur:divine, or a watchlist target's key). Resolve a user's wording to a key here first.

Latest recorded price for one item, with spread and confidence.

Reads the local database only. Returns found: false when the collector has not yet priced this item, which is not the same as the item being unsellable — check market_status.

OHLC price history for one item, for charting or trend analysis.

Reads rollup candles, so long windows stay fast. auto uses hourly candles up to 14 days and daily beyond that.

Items with the largest percentage price moves over a window.

Needs at least days + 1 days of collected history to be meaningful.

Search live trade listings right now, returning prices and whisper text.

Spends the shared GGG rate budget (600 searches / 6h), so prefer get_price when recorded data is good enough.

The returned whisper is the message to send in game. This server cannot send it: GGG exposes no trade-execution API, and automating in-game input violates the terms of service. A human completes the trade.

Pick the best current listing for an item and return a ready whisper.

Cross-checks the live ask against recorded history so an outlier price is flagged before you commit. Returns the whisper for a human to send; it does not contact anyone.

Show configured watchlists: what is scanned, how often, and priority.

Find currencies where the median bid exceeds the median ask.

Deliberately compares medians, not extremes. The cheapest ask on this endpoint is very often a fat-finger or a sold-but-still-listed order: the median listing is ~85 minutes old when GGG serves it. Divine was observed with asks of 100/188/260/300 against bids of 230/200/180/162/160 — the extremes cross by 130 exalted and imply free money, while the medians show an ordinary 26% spread. Screening on extremes would report a large opportunity in a market that has none.

A crossing that survives at the median is a genuine dislocation. Even then treat it as a lead, not a filled trade: both counterparties must be online, stock is finite, and these complete by whisper and a manual trade window, so the price can move before anyone replies.

stale_extremes reports books that cross only at the extremes, which is a staleness signal rather than an opportunity.

Value your own stash against collected prices.

Reads the account's public listings (no login) — the working path on PoE2, since the authenticated stash API is OAuth-only and closed. Only publicly-listed items are visible; set the account with stash_account in config or POE2MARKET_ACCOUNT.

Set refresh to pull fresh listings; otherwise the latest snapshot is returned.

Track how your stash's total value has changed over time.

Each snapshot is valued at the prices that applied when it was taken, so the series separates 'I acquired more' from 'what I hold got dearer'.

List what is actually in your stash: every item, with quantities.

Reads the most recent snapshot. Set group_stacks (the default) to aggregate the same item across tabs — 3 stacks of Exalted Orbs in three tabs become one row with the combined quantity, which is almost always the question being asked. Turn it off to see each physical stack and its tab.

Unpriced items (gear, maps, anything without a collected price) are included with a null value rather than dropped, so the list is a real inventory rather than only the part we happen to have priced.

Filters: search matches name or base type, tab matches a tab name, rarity is Normal/Magic/Rare/Unique/Currency/Gem.

Find profitable trade loops, e.g. chaos -> exalted -> divine -> chaos.

Pure analysis over already-collected direct pair rates: no API calls, so it is free to run as often as you like.

Only direct pair quotes can produce a cycle. Cross-rates synthesised through the base currency pay the spread on every leg, so a loop built from them always loses — profit appears only where the market's own chaos->divine rate has drifted from chaos->exalted->divine.

Each leg's rate has already paid its own spread and has had junk listings filtered out, so a gain above 1.0 is real profit rather than mid-price arithmetic. min_depth bounds how much can actually be pushed through the tightest leg.

As with all trades here: these complete by whisper and a manual trade window. Every counterparty must be online and willing. Treat results as leads, and note that a four-hop loop needs four separate humans to answer.

Fetch currency prices from poe.ninja right now, bypassing the schedule.

This is a hard refetch: it bypasses poe.ninja's CDN cache to reach the origin, rather than re-reading the cached response the hourly sweep uses. Worth calling when stale: true appears on a result, or after the machine has been asleep or offline. Note poe.ninja recomputes its own numbers about hourly, so a forced refetch guarantees the freshest published data — not necessarily different data.

Hits poe.ninja only; it does not touch GGG's trade API, so it cannot affect the player's in-game trade rate budget.