bevy-migration-0-17-to-0-18
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill provides documentation and refactoring examples for Rust code. It does not contain instructions to override system prompts, bypass safety filters, or exfiltrate internal agent state. As the skill facilitates processing external source code, it possesses an indirect prompt injection attack surface.
- Ingestion points: User-provided Rust source files and Cargo.toml project files (SKILL.md, references/cargo-feature-renames.md).
- Boundary markers: Absent; the skill relies on the agent's internal prompt separation for context management.
- Capability inventory: The agent using this skill typically has capabilities for file system modification (refactoring) and command-line execution (cargo commands).
- Sanitization: No explicit sanitization or validation of input source code is defined in the instructions.
- [DATA_EXFILTRATION]: No access to sensitive files (e.g., credentials, ssh keys) or unauthorized network requests were detected. The instructions focus purely on public API changes in the Bevy ecosystem.
- [REMOTE_CODE_EXECUTION]: The skill does not download or execute remote scripts. It provides guidance on updating local dependencies and cargo features using standard development practices.
Audit Metadata