The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a provided Python script (scripts/inspect-project.py) to analyze the local project environment. The script identifies Lincheck versions, API usage, and test configurations through static analysis of source and build files.
[INDIRECT_PROMPT_INJECTION]: The inspect-project.py script acts as a data ingestion point, reading local project files including Kotlin/Java source code, Maven POMs, and Gradle build scripts. While the script performs read-only regex matching, the resulting summaries and warnings are presented to the agent, creating a theoretical attack surface for indirect prompt injection via the analyzed codebase.
Ingestion points: Reads project files (.kt, .java, build.gradle, pom.xml, libs.versions.toml) via inspect-project.py.
Boundary markers: None explicitly implemented within the script output; the agent relies on structured JSON or plain text reporting.
Capability inventory: The script is limited to file system read operations and regex processing; it contains no file-write, network, or arbitrary command execution capabilities.
Sanitization: Uses standard encoding handling (errors="replace") during file reads.
[EXTERNAL_DOWNLOADS]: The skill documentation refers to and recommends fetching the latest Lincheck releases and documentation from official sources, including JetBrains' GitHub repositories and the Kotlin language guide.

lincheck