The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes an external file, template_spec.md, to determine the features and integrations of the generated template, which could be exploited to inject malicious instructions.
Ingestion points: template_spec.md is used as the primary input for generating the template structure and logic.
Boundary markers: The instructions do not specify any markers or system prompts to ignore instructions embedded within the specification file.
Capability inventory: The skill has the ability to write new directories and files to the templates/ folder and execute shell commands such as install, dev, and build.
Sanitization: No validation or sanitization of the specification content is mentioned before it is used to drive the generation and execution process.
[COMMAND_EXECUTION]: The workflow includes verification steps that require the agent to execute arbitrary commands within the context of the newly created template.
Evidence: The 'Verification' section (Step 3) instructs the agent to run install, dev, and build to confirm the template is functional, which involves executing the generated project's scripts.

workflow-template-seeder