The Agent Skills Directory

[SAFE]: No critical security vulnerabilities or malicious patterns were identified in the skill's instructions or logic.
[COMMAND_EXECUTION]: The skill executes commands using the infsh tool to perform image generation and information retrieval. This behavior is consistent with the skill's primary function and is restricted to the vendor's CLI namespace as defined in the allowed tools.
[EXTERNAL_DOWNLOADS]: The skill references external AI services and search providers through the vendor's CLI interface. These include well-known services such as Tavily and Exa for research, and Fal.ai or Bytedance for image generation.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests data from external sources via web search tools. This could allow maliciously crafted content from the web to influence the agent's behavior.
Ingestion points: Tool outputs from tavily/search-assistant and exa/answer are processed by the agent.
Boundary markers: No explicit delimiters or boundary instructions are provided to the agent to distinguish between search results and instructions.
Capability inventory: The skill can execute the infsh CLI tool to run various AI models and search applications.
Sanitization: There is no evidence of sanitization or filtering of the external data before it is incorporated into the prompt context.

landing-page-design