The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to perform repository analysis and search operations, such as identifying long functions or magic numbers using rg (ripgrep). It also instructs the agent to use git for commits and rollbacks.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it is designed to read and process untrusted external data (source code).
Ingestion points: Files are brought into the agent's context using the Read, Grep, and Glob tools during the identification and planning phases.
Boundary markers: The instructions do not define delimiters or provide 'ignore instructions' warnings when reading user-supplied code files.
Capability inventory: The skill has access to high-privilege tools including Bash (shell access), Edit, and Write (file system modification).
Sanitization: There is no evidence of sanitization or validation of the code content before it is processed by the agent.

loom-refactoring