The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides Bash and Python script templates (backup_social_media.sh and auto_archive.py) that use system commands such as git, mkdir, and cp. It also utilizes the GitHub CLI (gh) for creating issues.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and document content from external social media platforms, creating a vulnerability surface where the agent might process malicious instructions embedded in that data.
Ingestion points: Processes external content from social media platforms (e.g., Xiaohongshu) and local download folders in backup_social_media.sh and monitor_platform.py.
Boundary markers: No specific delimiters or warnings for the agent to ignore instructions within the ingested content are present in the provided templates.
Capability inventory: The skill utilizes file system operations (read/write), Git repository management, and GitHub API interactions.
Sanitization: The scripts do not include explicit sanitization or filtering logic for the content being archived.
[DYNAMIC_EXECUTION]: The Python script auto_archive.py uses subprocess.run to dynamically execute shell commands for file handling and Git version control based on runtime logic.

unisecurityguard-academic-whistleblower-archive