Early AD - True Positive list fetching

early-attack-detector-py/Dockerfile

@@ -24,6 +24,7 @@ ENV NODE_ENV=production
 LABEL "network.forta.settings.agent-logs.enable"="true"
 WORKDIR /app
 COPY ./src ./src
+COPY ./tp_list.csv ./
 COPY ./deployed_models ./deployed_models
 COPY package*.json ./
 COPY requirements.txt ./

early-attack-detector-py/README.md

@@ -79,6 +79,7 @@ The metadata for each alert will contain the following fields:
 - funding_labels: [optional] When there are potential fundings, the hash(es) to the funding label(s)
 - model_score: Model score. In case where the alert was raised due to the high precision model, the score will be from the high precision model.
 - model_threshold: Recall threshold that was being used in the model at the moment of raising the alert
+- known_past_attacker: [optional] When the transaction initiator is found to be a known past attacker, (i.e. associated with a public disclosed attack in the past) this value will be set to `"True"`.
 - oko_contract_explorer: Link to contract explorer
 
 Example of high precision model:

early-attack-detector-py/package.json

@@ -1,7 +1,7 @@
 {
   "name": "early-attack-detector-py-beta",
   "displayName": "Early Attack Detector Beta",
-  "version": "0.2.2",
+  "version": "0.2.4",
   "description": "Attack detector focused on detection during preparation",
   "longDescription": "This bot is designed to detect attacks during the contract deployment phase, before the attack is executed.",
   "repository": "https://github.com/forta-network/starter-kits/tree/main/early-attack-detector-py",

early-attack-detector-py/release.md

@@ -1,16 +1,26 @@
 # V0
 
+## V0.2.3
+
+- [beta] Added True Positive list and Hops beta bot.
+
 ## V0.2.2
-- [beta] Added squid, 1-hops, and SWFT Swap funding sources 
+
+- [beta] Added squid, 1-hops, and SWFT Swap funding sources
+- [prod] Added 1-hops, and SWFT Swap funding sources
 
 ## V0.2.1
+
 - [prod] Removed squid from funding sources temporarily
 
 ## V0.2.0
+
 - Updated model, new threshold. Updated docker version to be more lightweight.
 
 ## V0.1.0
+
 - Updated model
 
 ## V0.0.1
-- Initial version of the bot
+
+- Initial version of the bot

early-attack-detector-py/requirements.txt

@@ -3,7 +3,7 @@ setuptools>=61.3.1
 hexbytes>=0.2.2
 evmdasm>=0.1.10
 rlp>=2.0.1
-scikit-learn>=1.4.1
+scikit-learn==1.4.1.post1
 bot-alert-rate>=0.0.4
 web3==5.23.0
 imbalanced-learn>=0.12.0