The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The bootstrap section in SKILL.md downloads and executes a shell script from https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh by piping it directly to sh. This enables unverified third-party code to execute with the agent's privileges.\n- [COMMAND_EXECUTION]: The skill uses the Bash tool to interact with the system environment, specifically for installing the CLI, managing authentication status, and running image generation workflows.\n- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data and using it to construct prompts for the higgsfield CLI.\n
Ingestion points: Video topics, titles, scene requirements, and headlines are accepted as user input in SKILL.md.\n
Boundary markers: No boundary markers or isolation techniques are used to separate user-provided content from instructions in the generated prompts.\n
Capability inventory: The skill possesses the capability to execute shell commands and generate/modify images via the higgsfield CLI.\n
Sanitization: There is no evidence of validation or sanitization of the user-supplied strings before they are used in commands.

higgsfield-youtube-thumbnail