The Agent Skills Directory

[NO_CODE]: The skill package contains documentation, examples, and reference files in Markdown format but does not include any standalone scripts, binaries, or source code files.
[COMMAND_EXECUTION]: The skill instructs the agent to execute the dreamina CLI tool and docker commands to perform video generation tasks and verify environmental compatibility.
[EXTERNAL_DOWNLOADS]: The documentation references external content from the author's GitHub repository and provides instructions for installing related skills using the npx package runner.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of untrusted external data, including image, audio, and video files, alongside natural language prompts provided by the user.
Ingestion points: The skill uses file paths and prompt strings as arguments for CLI commands defined in SKILL.md and the examples/ directory.
Boundary markers: There are no explicit delimiters or specific instructions implemented within the command construction to isolate user-supplied prompts from tool instructions.
Capability inventory: The agent is granted the capability to execute shell commands (e.g., dreamina image2video, ls, docker run) to process the data.
Sanitization: The skill does not perform validation or sanitization of the content within the referenced media files or the prompts before they are passed to the CLI application.

dreamina-cli-image2video