TrojanSourceFinder is a Trojan Source vulnerability allows an attacker to make malicious code appear innocent. In general, the attacker tries to lure by passing his code off as a comment (visually). It is a serious threat because it concerns many languages. Projects with multiple “untrusted” sources could be concerned.
Install
go
Via go install
go install github.com/ariary/TrojanSourceFinder/cmd/tsfinder@latest
Make sure $GOPATH
is in your $PATH
From source
git clone https://github.com/ariary/TrojanSourceFinder
cd TrojanSourceFinder
make before.build
make build.tsfinder
If the command make build.tsfinder
failed, try:
env GOOS=target-OS GOARCH=target-architecture
go build -o tsfinder cmd/main.go
curl
From release
curl -lO -L https://github.com/ariary/TrojanSourceFinder/releases/latest/download/tsfinder && chmod +x tsfinder
Detect Trojan Source
> Help the detection of Trojan source for manual code review or with CI/CD pipelines (Unicode bidirectional characaters)
To detect Trojan source in file or directory <path>:
tsfinder [path]
Detect only in text file
> Source code files are likely text files. Withdraw them for scan could help to rule out false positives
tsfinder -t [path]
Add -v
help to see which file has been skipped by scan.
Trojan Source is not new and isn’t the only hazard. Another one is “Homoglyph”.(Kezako?)
tsfinder help detecting them with homoglyph
command:
tsfinder homoglyph [filename] [flags]
You could see if there is a sibling (ie word with same “skeleton”) for the homographs found in path
using the flag --sibling
:
tsfinder homoglyph [filename] –sibling [path]
Visualize Trojan Source
> Visualize how the code is really interpreted by machines/compiler
tsfinder is deliberately not very verbose. By default, it will only output if Trojan Source code has been detected. To have more verbosity and visualize the dangerous line add the flag -v
.
To better see where Trojan Sources were, you could enable colored output with -c
flag (also useful with directory scan):
tsfinder -c -v
Demo
Homoglyph
Playwright-MCP (Model Context Protocol) is a cutting-edge tool designed to bridge the gap between AI…
JBDev is a specialized development tool designed to streamline the creation and debugging of jailbreak…
The Kereva LLM Code Scanner is an innovative static analysis tool tailored for Python applications…
Nuclei-Templates-Labs is a dynamic and comprehensive repository designed for security researchers, learners, and organizations to…
SSH-Stealer and RunAs-Stealer are malicious tools designed to stealthily harvest SSH credentials, enabling attackers to…
Control flow flattening is a common obfuscation technique used by OLLVM (Obfuscator-LLVM) to transform executable…