neo-rust
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [DATA_EXPOSURE_AND_EXFILTRATION]: The 'SKILL.md' file contains hardcoded absolute local file paths (e.g., 'file:///Users/ben/Projects/neo-skills/...') used for loading reference documents. This exposes the author's local directory structure and usernames.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands including 'cargo check', 'cargo clippy', and 'cargo fmt'. While these are standard Rust development tools, they represent a surface for command execution.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection because it processes untrusted project files and uses them as arguments for toolchain commands.
- Ingestion points: The agent reads 'Cargo.toml' and '.rs' source files from the current project directory.
- Boundary markers: The instructions lack explicit delimiters or warnings to isolate untrusted code content from the agent's internal logic.
- Capability inventory: The skill utilizes file system access and executes shell commands ('cargo') that may trigger arbitrary code execution during the compilation or checking phase (e.g., via 'build.rs' scripts).
- Sanitization: No sanitization or validation of the untrusted code is performed before processing.
Audit Metadata