The Agent Skills Directory

[SAFE]: The skill provides instructions for thematic mapping and project architecture without any malicious patterns. It does not attempt to bypass safety filters or execute unauthorized commands.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data by reading existing project files and user descriptions to propose a structure. While this presents an attack surface where malicious content in a project file could attempt to influence the agent's output, the skill's capabilities are restricted to writing a structural map in markdown format at a specific path (./docs/structure.md). It does not have the capability to execute code or access sensitive credentials, mitigating the risk of such injections.
[COMMAND_EXECUTION]: The skill uses standard file system tools (Read, Grep, Glob, Write, Edit) for its intended purpose of project mapping. It does not invoke shells, compilers, or other high-risk command-line utilities.
[EXTERNAL_DOWNLOADS]: No external network operations or downloads were identified in the instructions or metadata.

cw-structure