The Agent Skills Directory

[PROMPT_INJECTION]: The skill uses clear, instructional language to guide the agent through its architectural mapping tasks. No attempts to override safety guidelines, extract system prompts, or bypass constraints were identified.
[DATA_EXFILTRATION]: The skill limits its operations to the local file system using 'Read' and 'Write' tools for architectural blueprints. It does not perform network operations or access sensitive credentials, environment variables, or private user files.
[REMOTE_CODE_EXECUTION]: No patterns involving external package installation, remote script downloads, or the execution of untrusted code were found. The skill generates text-based folder structures rather than executable scripts.
[COMMAND_EXECUTION]: The skill does not invoke shell commands, administrative privileges, or system-level persistence mechanisms. Its operations are restricted to documentation and directory mapping.
[OBFUSCATION]: All instructions and code examples are provided in plain, human-readable text. No Base64, hex encoding, or hidden Unicode characters were detected.

framework-adapter