The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data by reading project build configuration files to perform its analysis, creating a surface for potential injection if those files contain malicious instructions.
Ingestion points: The scripts/inspect-project.py script reads build.gradle, settings.gradle, and libs.versions.toml files from the target project directory.
Boundary markers: The script output provided to the agent lacks explicit boundary delimiters to separate the content of the analyzed files from the skill's instructions.
Capability inventory: The skill allows for local inspection script execution and provides instructions for modifying Gradle build files and running framework link tasks.
Sanitization: The script uses regular expressions and the standard tomllib parser to extract specific configuration data, which reduces the risk of executing arbitrary code contained within the analyzed files.
[COMMAND_EXECUTION]: The skill facilitates the execution of a local Python script (scripts/inspect-project.py) to analyze the environment. This script is limited to read-only operations on specific build-related files and does not perform network activities.
[EXTERNAL_DOWNLOADS]: The skill references and directs the agent to fetch documentation and plugins from official Touchlab SKIE domains (skie.touchlab.co). These are legitimate resources required for the skill's primary purpose.

skie