The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (tuist and git) using variables derived from user input, such as generation IDs and branch names. This interpolation requires safe parameter handling to prevent shell injection.
[INDIRECT_PROMPT_INJECTION]: The skill processes JSON data from tool outputs which may contain untrusted content from the project being analyzed, such as target names or commit messages.
Ingestion points: JSON output from tuist generate list --json and tuist generate show <id> --json.
Boundary markers: None specified in the instructions.
Capability inventory: Execution of tuist and git CLI tools as described in SKILL.md.
Sanitization: No validation or sanitization of the command output is performed before analysis.

compare-generations