The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown documentation and instructional templates. It does not include any executable scripts, binary files, or automated logic that performs operations on the host system.
[EXTERNAL_DOWNLOADS]: No external URLs, remote scripts, or third-party dependencies are referenced or downloaded.
[COMMAND_EXECUTION]: The skill does not execute any shell commands. While it mentions commands like pnpm install or npm test within markdown examples for project documentation, these are purely illustrative and are not triggered by the skill itself.
[DATA_EXFILTRATION]: The skill does not request network access or perform any data transmission. It focuses on guiding the user in writing better local instructions.
[PROMPT_INJECTION]: The content adheres to safety guidelines. It provides constructive guidance on security audits (e.g., checking for SQL injection or hardcoded secrets) and does not contain instructions to bypass filters or override system constraints.

prompt-engineering