The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use shell commands (mkdir, cp, ls) to prepare the directory structure, copy template files, and check for the presence of custom configuration overlays.
[INDIRECT_PROMPT_INJECTION]: The skill processes content from hiring/loop.md to generate the rubric dimensions and questions, creating a surface for potential instruction injection if the input file contains untrusted data.
Ingestion points: Reads the signal map from hiring/loop.md in step 1.
Boundary markers: The instructions do not define specific delimiters or warnings to ignore instructions that might be embedded within the signal map.
Capability inventory: The skill has the capability to write to the file system (hiring/rubrics/) and execute basic shell commands.
Sanitization: There is no mention of sanitizing the input from the signal map before using it to generate the JSON rubric.
[DYNAMIC_EXECUTION]: The skill implements an 'Overlay' pattern where the agent is instructed to dynamically load and prioritize instructions from a path defined by the $POWERSUIT_OVERLAY environment variable.

interview-rubric