The Agent Skills Directory

[PROMPT_INJECTION] (SAFE): No patterns of instruction override, safety bypass, or role-play injection were detected. The workflow is purely instructional and technical.
[DATA_EXFILTRATION] (SAFE): No network operations (curl, wget, fetch) or sensitive file path access (SSH keys, AWS creds) were found. Data usage is restricted to project-specific markdown files.
[REMOTE_CODE_EXECUTION] (SAFE): No remote script downloads or piped shell executions. The skill does not perform package installations or dynamic evaluation of external data.
[COMMAND_EXECUTION] (SAFE): No execution of system commands or shell scripts. The skill focuses on generating architectural documentation.
[OBFUSCATION] (SAFE): No Base64 encoding, zero-width characters, or homoglyph-based obfuscation detected in the text or metadata.
[INDIRECT_PROMPT_INJECTION] (SAFE): While the skill ingests project-related markdown files (e.g., hooks_catalog.md), the risk is minimal as the outputs are restricted to document generation and internal memory updates without executable side effects.

react-expert