The Agent Skills Directory

[PROMPT_INJECTION]: No instructions were found that attempt to bypass AI safety guidelines, override system behavior, or extract system prompts. The skill uses structured checklists to guide agent behavior within its intended scope.
[DATA_EXFILTRATION]: The skill does not contain any commands to access sensitive local files (like SSH keys or AWS credentials) or perform unauthorized network requests. It explicitly instructs the agent to check for secret exposure and logging in the code being reviewed, which is a security benefit.
[REMOTE_CODE_EXECUTION]: No remote script downloads (e.g., curl|bash), package installations, or dynamic code execution patterns (e.g., eval or subprocess calls) were detected in the instructions or scripts.
[SECURE_PRACTICES]: The skill includes a 'SECURE Checklist' that mandates input validation, prevention of path traversal, and secure handling of external data, demonstrating a high focus on software security best practices.
[OBFUSCATION]: All instructions and metadata are in clear text. No Base64, zero-width characters, homoglyphs, or other obfuscation techniques were identified.

code-foundations