ratatui-tui
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The
tui-review.jsworkflow script performs automated code reviews by ingesting source code from the user's workspace. - Ingestion points: The script reads files from the directory specified in the
pathargument (e.g.,src/) to identify TUI-specific issues. - Boundary markers: While the script does not use explicit delimiters around the ingested code, it utilizes a task-specific prompt structure for different TUI dimensions (architecture, styling, etc.).
- Capability inventory: The review process involves file-reading and analysis capabilities within the AI agent's environment.
- Sanitization: The script implements an adversarial verification phase where a second agent attempt to refute each finding, serving as a mitigation against potential manipulation or false positives.
- [COMMAND_EXECUTION]: The skill documentation provides instructions for standard terminal-based development tasks.
- Evidence: Recommended commands include project initialization via
cargo generate, local file management usingcp, and application execution withcargo run. - [EXTERNAL_DOWNLOADS]: The documentation references established third-party tools and community libraries.
- Evidence: The skill guides users to install the official
cargo-generatetool and suggests incorporating thetui-shimmeranimation crate from its public GitHub repository. These are well-known resources in the Rust ecosystem.
Audit Metadata