The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external performance data to iterate on ad creative, which creates a potential surface for instructions embedded in that data to influence the agent.
Ingestion points: The skill processes user-provided performance data in CSV or text format as described in SKILL.md (Mode 2: Iterate from Performance Data).
Boundary markers: There are no explicit instructions for the agent to use delimiters or ignore embedded natural language instructions within the ingested data.
Capability inventory: The skill's primary capabilities are text generation and documentation. It does not include dangerous runtime capabilities such as arbitrary file writing or system command execution based on the ingested data.
Sanitization: No specific validation or sanitization routines are defined for the input data structure.
[EXTERNAL_DOWNLOADS]: The skill's reference documentation provides instructions for downloading and using various third-party tools and services.
references/generative-tools.md includes curl examples targeting official APIs for Google Gemini and ElevenLabs.
The documentation provides instructions for the user to clone an external GitHub repository (github.com/jamiepine/voicebox.git) and install packages using npx (create-video).
[COMMAND_EXECUTION]: The skill provides example CLI commands intended for the user to execute for media processing and campaign management.
Includes ffmpeg commands for merging audio and video files.
Includes examples for running local Node.js scripts (e.g., node tools/clis/google-ads.js) to fetch advertising reports.
Includes remotion CLI commands for rendering video ad templates.

ad-creative