The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines several templates for logo generation that interpolate user-provided parameters (such as [subject], [style], and [color]) directly into shell commands. This creates a vulnerability surface where maliciously crafted user input could potentially break out of the command string.
Ingestion points: User-controlled fields within the prompt templates in SKILL.md.
Boundary markers: Absent. No specific delimiters or instructions are provided to the agent to treat these inputs as literal text only.
Capability inventory: The skill uses the Bash tool to execute belt commands.
Sanitization: Absent. The instructions do not specify any validation or escaping requirements for the user-supplied strings before they are placed into the CLI arguments.
[EXTERNAL_DOWNLOADS]: The skill directs users to install an external command-line interface (belt-sh/cli) and references installation scripts from a remote GitHub repository. These downloads are required for the primary functionality of the skill.
[COMMAND_EXECUTION]: The skill provides numerous examples of shell commands involving the belt CLI tool, including a loop structure for batch processing images. These commands are executed via the Bash tool.

logo-design-guide