The Agent Skills Directory

[PROMPT_INJECTION]: The skill provides operational guidelines for delegating tasks. It does not contain any instructions intended to bypass safety guardrails, disregard previous rules, or extract system prompts. The role definitions are purely functional for task management.
[DATA_EXFILTRATION]: No sensitive file paths, credential markers, or network communication patterns (like curl, wget, or fetch) were found. The skill does not interact with external servers.
[REMOTE_CODE_EXECUTION]: The skill does not include any package manager instructions (npm, pip) or remote script execution patterns. It relies entirely on static markdown instructions.
[OBFUSCATION]: All content in the skill files and reference documents is provided in clear-text markdown and JSON. No Base64, hex encoding, or zero-width character obfuscation was detected.
[COMMAND_EXECUTION]: There are no shell commands or system-level calls identified in the instructions or reference materials.

roles