text-to-lottie
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes shell commands for environment management and file validation. It provides instructions to run a local development server using
npm run devand uses anode -eone-liner inSKILL.mdto verify the integrity of Lottie JSON files after generation or modification. - [EXTERNAL_DOWNLOADS]: The skill provides instructions in
references/player-contract.mdfor setting up the local player environment by downloading a template from thediffusionstudio/lottierepository usingnpx degitand installing necessary packages vianpm install. These actions are performed from known sources within the animation development ecosystem. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and manipulate external data, such as user-provided SVG assets and existing Lottie JSON files. While these inputs could theoretically be used to influence the agent's behavior, the skill includes specific validation steps and structured processing rules to mitigate such risks.
- Ingestion points: Reads
lottie.jsonfiles and SVG source data from the project directory as part of its core workflow. - Boundary markers: Includes instructions to validate JSON syntax using local commands and verify results against official player behavior.
- Capability inventory: Capabilities include file system writes (writing Lottie JSON) and local command execution (node validation, npm commands).
- Sanitization: Relies on JSON parsing and Skia Skottie renderer compatibility for processing and rendering inputs.
Audit Metadata