The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external search providers (Tavily and Exa) to assist in newsletter curation. This creates an attack surface where untrusted content from the web could influence the agent's behavior.
Ingestion points: Search results retrieved via belt app run tavily/search-assistant and belt app run exa/search (referenced in SKILL.md).
Boundary markers: The instructions do not define explicit delimiters or instructions to ignore embedded prompts within the retrieved search results.
Capability inventory: The skill utilizes the belt CLI for network operations and content generation. Bash access is restricted to the belt command only via the allowed-tools configuration.
Sanitization: There is no evidence of explicit sanitization or validation of the content retrieved from external search engines.
[EXTERNAL_DOWNLOADS]: The skill references installation guidelines and scripts hosted on the official GitHub repository of the author (inference-sh) at https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md.
[COMMAND_EXECUTION]: The skill makes use of the belt CLI for various tasks, including social post creation (x/post-create) and image generation (infsh/html-to-image). Tool execution is properly scoped to the belt namespace.

newsletter-curation