The Agent Skills Directory

[SAFE]: The skill provides documentation and implementation details for a legitimate messaging service. All identified external resources and API endpoints belong to the verified author's infrastructure.- [EXTERNAL_DOWNLOADS]: The skill includes a reference to a GitHub repository (github.com/kudosity/ai-agent-examples) for a working example. This is a trusted source belonging to the skill author.- [CREDENTIALS_UNSAFE]: The skill correctly uses a placeholder ({KUDOSITY_API_KEY}) for authentication, following best practices for secret management.- [DATA_EXFILTRATION]: Communication is directed to the vendor's official API domain (api.transmitmessage.com). No unauthorized data exfiltration patterns were observed.- [INDIRECT_PROMPT_INJECTION]: The skill facilitates interaction with external API responses and webhooks. While this creates a potential attack surface for indirect prompt injection, it is managed by the agent's internal safety protocols and the skill itself does not contain exploitable logic.

kudosity-whatsapp