The Agent Skills Directory

[SAFE]: No security issues detected. The skill is designed to facilitate the configuration and maintenance of the detekt static analysis tool within Kotlin projects. Its operations are local and consistent with standard developer workflows.
[INDIRECT_PROMPT_INJECTION]: The skill reads project-specific configuration files (such as build.gradle and libs.versions.toml) to determine the current project state. This presents a typical ingestion surface for untrusted data, though the risk is negligible as it is restricted to specific build files and the agent's actions are focused on tool configuration.
Ingestion points: scripts/inspect-project.py reads and parses build-related files.
Boundary markers: Absent; the agent relies on the structured data returned by the helper script.
Capability inventory: The skill permits the agent to execute local Gradle commands and modify configuration files as requested by the user.
Sanitization: The inspection script uses specific, narrow regular expressions to extract only intended version strings and plugin identifiers, minimizing the risk of processing malicious instructions embedded in the project files.

detekt