The Agent Skills Directory

[SAFE]: The skill's primary functions—binary analysis and project sanitization—are performed locally using standard system utilities such as strings, nm, otool, sed, and find.
[COMMAND_EXECUTION]: The provided bash scripts (analyze_target.sh and sanitize_project.sh) execute shell commands to analyze binary properties and perform string replacements. These actions are transparent and aligned with the skill's documented purpose. Safety checks in sanitize_project.sh prevent accidental execution on critical system directories like / or the user's home directory.
[DATA_EXFILTRATION]: Analysis of the scripts and documentation revealed no network requests, remote connections, or data exfiltration attempts. All processing is confined to the local filesystem.
[PROMPT_INJECTION]: The skill instructions and reference materials do not contain patterns designed to manipulate agent behavior, override safety filters, or extract system prompts.
[CREDENTIALS_UNSAFE]: There are no hardcoded credentials, API keys, or instructions that would lead to the exposure of sensitive secrets or local credential stores.

project-sanitizer