The Agent Skills Directory

[SAFE]: The skill focuses on promoting industry-standard best practices in TypeScript, particularly around security-sensitive areas like secret management and input validation. It explicitly prohibits hardcoded credentials and non-null assertions on external data.
[SAFE]: The included Node.js scripts in the evals/ directory are intended for internal validation and testing. These scripts perform benign file system operations within the skill's own directory to check for structural invariants and run evaluation scenarios.
[SAFE]: The 'adversarial prompts' found in the evals/ folder are used for behavioral testing of the agent's logic and do not constitute an attempt to bypass the agent's core safety guidelines.
[SAFE]: The skill uses clear, well-documented Markdown and TypeScript files. No evidence of obfuscation, hidden URLs, or suspicious dynamic code generation was found.
[SAFE]: All external references in the documentation point to well-known technological patterns or reputable open-source projects for educational purposes, posing no security risk.

typescript-skills