The Agent Skills Directory

[NO_CODE]: The skill is comprised entirely of markdown-based instructions for the AI agent and does not contain any executable code, scripts, or binaries.
[SAFE]: No malicious patterns, such as hardcoded credentials, network exfiltration, or obfuscated commands, were identified.
[DATA_EXFILTRATION]: The skill includes a 'Rules' section that explicitly forbids reading file contents, allowing only file paths to be collected as references, which protects against the exposure of sensitive local data.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection because it saves user-provided task descriptions into markdown files intended for future AI sessions; however, this is an inherent part of its documentation function and is managed by a required user confirmation step before any file is written.

define-task