The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill retrieves data from external web sources and user-supplied files, then uses this content to construct a storyboard and generate Python code for video rendering.
Ingestion points: SKILL.md Step 2 describes gathering proof from user sources and web searches.
Boundary markers: No specific delimiters or safety instructions are defined to prevent the agent from following malicious instructions hidden within the ingested data during the script generation phase.
Capability inventory: The skill environment allows for shell command execution, Python script generation, and media rendering using Manim.
Sanitization: There are no explicit instructions for sanitizing or escaping the external data before it is written into the scenes.py script.
[DYNAMIC_EXECUTION]: The skill generates and runs Python code (scenes.py) at runtime to produce video frames.
Evidence: SKILL.md Step 5 instructs the agent to write and execute a scenes.py file using the Manim engine.
[EXTERNAL_DOWNLOADS]: The skill installs dependencies from the author's GitHub repository.
Evidence: SKILL.md provides an installation command using npx targeting Pesty-Marketing/agent-skills. This reference to the vendor's own repository is documented neutrally.

pesty-explainer