The Agent Skills Directory

[SAFE]: The skill provides educational content and technical implementation details for live chat in e-commerce contexts without any malicious patterns.
[EXTERNAL_DOWNLOADS]: References official and well-known services (Shopify Inbox, Tidio, Gorgias, LiveChat) for e-commerce integrations, which are considered safe per established rules.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns or untrusted script downloads were found.
[COMMAND_EXECUTION]: No shell command execution or risky system calls are present.
[DATA_EXFILTRATION]: No patterns for accessing sensitive files or exfiltrating user data to external servers were identified.
[PROMPT_INJECTION]: No attempts to override agent behavior or bypass safety guidelines were found in the instructions.
[OBFUSCATION]: No obfuscated content, Base64-encoded commands, or hidden URLs were detected.
[PRIVILEGE_ESCALATION]: No requests for elevated permissions or administrative access were identified.
[PERSISTENCE]: No mechanisms for maintaining access across sessions or modifying system start-up files were found.
[METADATA_POISONING]: The frontmatter and metadata are consistent with the skill's stated purpose and do not contain deceptive content.
[INDIRECT_PROMPT_INJECTION]: While the skill processes chat messages, it includes architectural advice on handling untrusted data securely (e.g., sanitization) and does not exhibit exploitable capability tiers in its static code examples.
[TIME_DELAYED_CONDITIONAL]: No time-gated or conditional logic targeting security-sensitive operations was found.
[DYNAMIC_EXECUTION]: The provided code snippets use standard WebSockets and React patterns without unsafe dynamic execution like eval() or runtime compilation.
[DYNAMIC_CONTEXT_INJECTION]: No use of the !command syntax was found in the SKILL.md file.

live-chat-commerce