BlueCLMarket observations
The backend collects public economy observations and stores historical snapshots. Displayed prices are observations, not guaranteed execution prices.
POE2 Exchange / Game economy & data
Read the market. Find the pattern.
POE2 Exchange brings item prices, historical charts, exchange volume and time-of-week patterns into one dashboard. Short summaries help players interpret the observations beside the charts.
Development complete
Three steps
From the first step to a useful result.
- 01
Choose your league
Open the dashboard and select the league you play in.
- 02
Find an item
Search for an item, then compare the price chart and exchange information.
- 03
Read the context
Review weekday/hour patterns and sample counts alongside the summary.
PRODUCT / IN VIEW
See the product.
Explore the interface, then open the service, source code or demo using the links above. Select an image to see it in full.
Getting started
Start with the right setup.
Access & setup
- Use the live browser dashboard; no desktop installation is needed. The methodology and guide pages explain how observations are collected and compared.
Permissions & data
Understand what the workflow uses.
AI summary input
The summary request sends the selected item name and computed statistics to the model provider. The output explains recorded observations.
This is product usage guidance. The BlueCL website privacy notice covers this company website; consult the service’s own documentation for its deployed data handling. Website privacy ↗
Behind the product
How the product works.
Players need to compare price observations and exchange activity across multiple sources. A chart can show movement, but understanding the sampling period, relative price patterns and volume concentration still takes work.
01Public data sourcesPrice observations + official exchange volume
02Local statisticsSQLite history · sample counts · time patterns
03Summary generationExplain the supplied observations
04Dashboard summaryShort Korean explanation beside the data
Model integration
After aggregation, /api/market-summary sends the item name and computed statistics to Claude Haiku 4.5 for a short Korean explanation. The model is configurable through CLAUDE_MODEL. Collection and chart calculations use application code.
Implementation details
The implementation prepares a statistical summary first. If the Claude key is absent or the API request fails, that summary is returned instead. The prompt asks for observed tendencies and sample limits and excludes future-price predictions. Source price snapshots are kept separate from official exchange-volume records.
Further improvements
Evaluate summary faithfulness against the supplied statistics, improve data freshness visibility, and measure summary latency and API cost.
When you need a hand
Frequently asked questions.
Are the prices live trade prices?
They are sampled reference observations. Check freshness, source and liquidity context before interpreting them.
Can I compare patterns across days?
The time-pattern view compares observations relative to each day’s median. Read the sample counts and observation period with it.
What if the model API is unavailable?
The implementation returns its statistical summary as a fallback.
Product and source information updated: October 8, 2026