Bot intent classifier
Read a client's recent requests and name what it's after and how it crawls.
Hand Jev the last few dozen paths a client requested. It names the target (prices, articles, accounts, stock, search, media, your API) and the pattern (walking pagination, enumerating IDs, following the sitemap, hammering search, trying logins), so you can block, slow down or allow by intent instead of by IP address.
Try it live
Real calls to Jev. Pick an example or paste your own.
Pick an example or paste your own input, then run it.
Under the hood
Every question and rule, in the open. Fork it to change any of them.
What it asks
Reads requests and client, then asks 4 questions in one call.
- Q1AfterPick one
What is the client mostly after, judging by the
requests?PricesContentAccountsStockSearchMediaAPIUnclear - Q2PatternPick one
How do the
requestsmove through the site?Walks pagesEnumerates IDsFull crawlSearch sweepLogin attemptsBrowsingOther - Q3AutomatedYes / no
Do the
requests, together with what's known about theclient, look automated rather than a person in a browser (no page assets, a machine-regular order, or far more pages than a person reads)? - Q4AggressiveYes / no
Do the
requestslook like an attempt to copy the site's data wholesale, or to get into accounts?
How it decides
Rules run in code, top to bottom. The first one that matches gives the verdict.
- 1If Pattern is Login attemptsBlock
- 2If Automated ≥ 0.7 and Aggressive ≥ 0.7Block
- 3If Automated ≥ 0.6Slow down
- ·DefaultAllow
Call it from your code
Every template runs as-is with an API key, no setup: POST /api/v1/judges/bot-intent-classifier/run.
curl -X POST https://tryevally.com/api/v1/judges/bot-intent-classifier/run \ -H "Authorization: Bearer $EVALLY_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "input": { "requests": "GET /products?page=1\nGET /products?page=2\nGET /products?page=3\nGET /products?page=4\nGET /products/10431\nGET /products/10432\nGET /products/10433\nGET /api/price?sku=10431\nGET /api/price?sku=10432\nGET /api/price?sku=10433", "client": "python-requests/2.32, data-center network" }}'Seen in the wild
Community builds on Jev that use this pattern. Not affiliated with Evally.
More in Trust & safety
- Trust & safety
Community moderator
Approve, review or remove posts against your own community rules, with severity.
ApproveSend to moderatorsRemove+18 questions - Trust & safety
Spam detector
Separate real messages from promotion, bait and bots, and only delete what's clear.
Not spamMaybe spamSpam5 questions - Trust & safety
Scam message detector
Paste an SMS, email or DM and get the probability it's a scam, and why.
Looks safeSuspiciousScam7 questions - Trust & safety
Fake review detector
Flag fake and incentivized reviews, and route real defect reports to the right team.
GenuineSuspiciousLikely fake+16 questions - Trust & safety
Marketplace listing check
Catch prohibited items, counterfeits and off-platform deals before a listing goes live.
ApproveReviewReject5 questions - Trust & safety
Conversation red-flag detector
Paste a chat and spot love-bombing, money requests, pressure and inconsistent stories.
Looks fineBe carefulRed flags5 questions