What If Keepa Could Understand What You're Actually Looking For?

Agellic is a conversational AI interface for Keepa, the Amazon product data and price tracking service. It lets Amazon sellers search the Keepa Product Finder database using natural language instead of manually configuring hundreds of filter parameters. You describe what you are looking for in plain English, like "find low-competition supplements under $30 with rising sales rank," and Agellic translates that into an optimized Keepa query, returns results, and presents calculated analytics so you can evaluate opportunities quickly.

Yes. Agellic uses a bring-your-own-key (BYOK) model. You connect your personal Keepa API key, and all queries run against your own Keepa token balance. This means you retain full control over your API usage and costs. Agellic includes smart token management that shows your remaining balance and optimizes query efficiency to reduce unnecessary token consumption.

Your Keepa API key is encrypted with AES-256-GCM before it touches our database. It is never logged or stored in plaintext. Cached Keepa data is namespaced to your API key, meaning other users cannot access your cached results. Your saved searches, product lists, and conversation history are not shared between users, not sold, and not used to build aggregate features. No advertising trackers or cross-site tracking pixels. We use privacy-focused product analytics.

Keepa Product Finder is extremely powerful but has a steep learning curve. Hundreds of filter parameters, cryptic field names, and no guidance on which combinations produce useful results. Agellic removes that friction. You describe your criteria conversationally, and the AI agent constructs the right Keepa query, iterates using Search Insights to refine results, and calculates advanced analytics that Keepa does not provide natively. These computed metrics give you a faster, deeper read on product opportunities without needing to master every Keepa filter.

Agellic computes metrics that go well beyond raw Keepa data. Core analytics include demand stability scoring, Amazon out-of-stock pattern detection (intermittent, seasonal, or chronic), price trend analysis with promo spike filtering, and seller concentration using HHI (Herfindahl-Hirschman Index). Advanced analytics include seasonality detection with phase timing (peak and trough month predictions), price velocity and acceleration tracking, FBA vs FBM concentration and price differentials, sawtooth repricing pattern detection, race-to-bottom early warning with signal-family aggregation, stock depth analysis with depletion rate and inventory turnover benchmarks, Buy Box volatility scoring, effective competition measurement, seller cliff detection, review purge detection, price position percentile scoring, and composite risk assessments including IP risk and race-to-bottom early warning scores. All analytics are calculated server-side from Keepa historical data. The AI agent never sees raw API responses.

All Amazon marketplaces available through the Keepa API, including US, UK, Germany, France, Italy, Spain, Canada, Japan, India, Mexico, Brazil, and others. Agellic supports cross-market searches, so you can compare product performance across multiple Amazon domains in a single query. This is useful for identifying international arbitrage opportunities or evaluating market expansion.

Yes. You can save any search to re-run later with a single click, and organize products into custom lists for ongoing tracking and comparative analysis. Saved searches remember your exact criteria so you can monitor how results change over time. Product lists let you group ASINs across different searches and categories for side-by-side evaluation.

Keepa API access is metered by tokens. Each query consumes tokens based on the number of results and the data fields requested. For example, a Finder search costs 11 tokens, and a detailed product lookup costs 9 tokens per ASIN. Agellic requires a minimum 20 tokens per minute Keepa plan. The standard 1 token per minute plan is too slow for conversational use, as a single search would take 11 minutes to complete. Agellic includes a smart token management system that reserves tokens for interactive use so background jobs do not starve your live sessions, caches product data for 24 hours to avoid repeat costs, and lets you check your balance at any time during a conversation.