The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Maven (mvn) commands, such as mvn -q javadoc:jar -DskipTests for linting and mvn -q spotless:apply for code formatting. These are standard development operations restricted to the local project environment.
[INDIRECT_PROMPT_INJECTION]: The skill processes local project files, which constitutes a potential attack surface for indirect prompt injection from malicious code content.
Ingestion points: Project configuration files (pom.xml, build.gradle) and Java source files (.java) are read from the local repository as part of the detection and linting workflow.
Boundary markers: Absent. The instructions do not specify the use of delimiters or specific warnings to ignore instructions that might be embedded in the processed files.
Capability inventory: The skill possesses the capability to execute shell commands via the mvn build tool as defined in the toolchain.md reference file.
Sanitization: Absent. No specific sanitization or filtering logic is described for the content read from local files before it is processed by the agent.

codeguard-java