The Agent Skills Directory

[PROMPT_INJECTION]: No malicious instructions or attempts to bypass AI safety guardrails were detected. The content consists of legitimate coding patterns and technical guidance.- [DATA_EXFILTRATION]: No code or instructions for unauthorized data access or network exfiltration were found. The snippets focus on standard application logic such as UI notifications and API response handling.- [REMOTE_CODE_EXECUTION]: The skill does not contain patterns for downloading or executing remote scripts. It uses standard library imports and functional wrappers.- [COMMAND_EXECUTION]: No dangerous shell commands or privilege escalation attempts were identified. File system operations mentioned are synchronous checks common in development environments.- [EXTERNAL_DOWNLOADS]: The skill references vendor-specific libraries ('wellcrafted' and '@epicenter/ui') that align with the author's identity. No suspicious third-party dependencies or untrusted sources were found.- [CREDENTIALS_UNSAFE]: No hardcoded secrets, API keys, or private tokens were discovered in the code examples or metadata.

error-handling