> For clean Markdown of this page, append .md to its URL. For the complete documentation index, see https://www.superagent.sh/llms.txt.


Review where untrusted data reaches components that interpret commands, queries, or other instructions.

# Interpreter injection in source code


[← Back to attack scenarios](https://www.superagent.sh/docs/products/red-teaming/attack-scenarios)

A service may pass data from outside the system to an interpreter. This scenario examines whether the code keeps that data separate from the instructions the interpreter is meant to follow.

It applies to repositories that work with interpreters and accept untrusted data. A finding identifies the source code boundary where input acquired the ability to affect execution.

The app lists this scenario as **Turn untrusted input into interpreter execution** (`REPO-07`) in [Repository Red Teaming](https://www.superagent.sh/docs/products/red-teaming/repositories).

## Standards and guidance

[OWASP A05:2025 Injection](https://top10.owasp.org/2025/A05_2025-Injection/)

---
Source: https://www.superagent.sh/docs/products/red-teaming/attack-scenarios/repository-interpreter-injection
Index: https://www.superagent.sh/llms.txt
