The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project data and build outputs to perform its primary function, creating an attack surface where malicious content could influence the agent's logic.
Ingestion points: The skill reads project configuration files (e.g., pom.xml, build.gradle, build.gradle.kts) and build logs (e.g., native-build.log) using tools like Grep, Read, and Bash to detect frameworks and troubleshoot errors.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present when the agent is directed to analyze these files.
Capability inventory: The skill possesses significant capabilities, including executing shell commands (Bash), writing files (Write, Edit), and running the compiled native executable (e.g., ./target/<app-name>).
Sanitization: There is no evidence of sanitization or strict schema validation for the data ingested from the project environment or build logs.

graalvm-native-image