The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides the command curl -fsSL https://cli.inference.sh | sh, which downloads and executes a script from a remote URL. This pattern allows for arbitrary code execution and is highly discouraged because the script content is not verified before it runs.
[EXTERNAL_DOWNLOADS]: The skill requires downloading content and tools from https://cli.inference.sh and https://dist.inference.sh, which are not recognized as trusted sources.
[COMMAND_EXECUTION]: The skill uses the Bash tool to run the infsh CLI and its associated applications, such as tavily/search-assistant and exa/search. These tools make external network connections to retrieve data.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection based on the following evidence:
Ingestion points: Data from search assistants (tavily/search-assistant, exa/search) is ingested into the agent context (file: SKILL.md).
Boundary markers: There are no boundary markers or instructions to ignore embedded commands in the external data.
Capability inventory: The agent can execute commands via the infsh tool using the Bash capability.
Sanitization: There is no validation or sanitization of content retrieved from the web before it is processed by the agent.

newsletter-curation