The Agent Skills Directory

[DATA_EXPOSURE] (SAFE): The scripts perform local file operations to package files into a ZIP format and validate metadata. No access to sensitive system paths (e.g., ~/.ssh, ~/.aws) or hardcoded credentials was found.
[DYNAMIC_EXECUTION] (SAFE): The validation script uses yaml.safe_load() to process frontmatter, which is the secure method for parsing YAML and prevents arbitrary code execution during deserialization. No use of eval(), exec(), or unsafe subprocess calls on untrusted data was detected.
[PROMPT_INJECTION] (SAFE): The provided markdown documentation files contain instructional patterns for guiding agent behavior. These are standard formatting and workflow guides and do not contain bypass instructions, role-play jailbreaks, or attempts to extract system prompts.
[EXTERNAL_DOWNLOADS] (SAFE): No network operations, external downloads, or remote script execution patterns (e.g., curl | bash) were identified in the scripts.

skill-creator