The Agent Skills Directory

[PROMPT_INJECTION]: Analysis of the instructions confirms the absence of patterns designed to override agent behavior, bypass safety guardrails, or extract system prompts.- [DATA_EXFILTRATION]: No network operations, sensitive file path access, or credential exposures were identified. The content is purely instructional and does not interact with the host system or external network.- [REMOTE_CODE_EXECUTION]: The skill does not contain instructions for downloading or executing remote code, nor does it include package installation commands from unverified sources.- [COMMAND_EXECUTION]: There are no shell commands, subprocess calls, or system-level operations present in the documentation or code snippets.- [OBFUSCATION]: No evidence of Base64 encoding, zero-width characters, homoglyphs, or other techniques used to hide malicious intent was found.- [PRIVILEGE_ESCALATION]: The skill does not request elevated permissions, use sudo, or attempt to modify system configuration files.- [PERSISTENCE]: No mechanisms for establishing persistence, such as modifying shell profiles or scheduled tasks, were detected.- [DYNAMIC_EXECUTION]: The code snippets provided are static examples for frontend development and do not involve runtime code generation, unsafe deserialization, or library injection.- [DYNAMIC_CONTEXT_INJECTION]: The skill does not utilize dynamic context triggers or shell command execution at load time.

accessibility-interaction-patterns