The Agent Skills Directory

[SAFE]: The skill's primary function is to conduct a depth-first interview to resolve decision branches in a plan or design. Its logic is based on interacting with the user and analyzing existing project files.
[COMMAND_EXECUTION]: The skill utilizes a restricted set of tools (Read, Grep, Glob) as specified in its configuration. These tools are used for static analysis of the codebase to answer questions internally before prompting the user, posing no risk of arbitrary command execution.
[DATA_EXPOSURE_&_EXFILTRATION]: No network operations, external URLs, or hardcoded credentials were detected. The skill operates locally on the provided context.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests data from local files using Read and Grep, it lacks the capability to write files or execute code (eval/exec), which significantly limits the risk of instructions embedded in analyzed files influencing the agent beyond the scope of the interview.

grill-me