The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected. The skill acts as a meta-tool for developing and testing other skills.
[PROMPT_INJECTION]: The documentation includes instructions on using persuasion techniques such as "Authority" and "Commitment" to ensure agents follow strict rules (e.g., TDD discipline). These techniques use imperative language ("YOU MUST", "No exceptions") to harden instructions against accidental non-compliance during complex tasks, rather than to bypass safety guardrails.
[COMMAND_EXECUTION]: The utility script render-graphs.js executes the system command dot to generate SVG diagrams from markdown content. The script uses standard input for data passing, which is a safe practice that prevents typical shell injection vulnerabilities.

writing-skills