The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process and refactor external code, which is a potential vector for indirect prompt injection if the source code contains malicious instructions.
Ingestion points: The agent reads source code files from the user's workspace to identify simplification opportunities.
Boundary markers: The skill does not specify particular delimiters or escaping mechanisms for the code being read, though it emphasizes deep comprehension (Chesterton's Fence) before modification.
Capability inventory: The skill assumes the agent has read/write access to files within the project scope.
Sanitization: No explicit sanitization or filtering of the input code content is defined.
[SAFE]: The skill consists entirely of markdown instructions and illustrative code snippets. No executable scripts, binaries, or active configuration files are included.
[SAFE]: No evidence of credential theft, sensitive file exfiltration, or unauthorized network operations was found. The skill references a trusted organization's repository for additional context, which is considered safe behavior.
[SAFE]: No obfuscation, prompt injection bypasses, or persistence mechanisms were detected in the skill content.

code-simplification