make-short
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the agent to execute a variety of local scripts and system commands (npm, node, python, ffmpeg) to manage the video lifecycle. This involves significant interaction with the project environment and the local shell.
- [DYNAMIC_EXECUTION]: The workflow involves the automated generation of TypeScript React (TSX) components for Remotion and timing configuration files which are then executed during the rendering process. While based on templates, the generation and execution of code at runtime is a core capability.
- [INDIRECT_PROMPT_INJECTION]: The process begins with a user-provided "topic" that influences the generated script, configuration, and potentially file paths, creating a surface for adversarial manipulation.
- Ingestion points: User-provided "topic" in the initial prompt used to generate
script.mdandbeats.jsonin Stage 1. - Boundary markers: Absent; the instructions do not define specific delimiters to isolate user input from the agent's logic during the creation of production artifacts.
- Capability inventory: Shell command execution (Node.js, Python, FFmpeg), local file system writes (scripts, code, assets), and network connectivity for voice generation services.
- Sanitization: Absent; no explicit validation or sanitization steps are defined for user-provided strings before they are used in generated files or as command-line context.
Audit Metadata