rust-engineer
Fail
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: CRITICALINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied Rust code and project files, creating a surface for indirect prompt injection attacks.\n
- Ingestion points: Reads and analyzes source code, Cargo.toml, and project directories.\n
- Boundary markers: The instructions lack specific delimiters or 'ignore' directives to separate user data from agent instructions.\n
- Capability inventory: The agent is authorized to run cargo test, cargo clippy, and cargo bench, as well as perform file system modifications.\n
- Sanitization: No input validation or sanitization of ingested code is specified.\n- [COMMAND_EXECUTION]: The skill workflow involves executing several Rust-specific toolchains that can run arbitrary code.\n
- Evidence: cargo clippy, cargo fmt, cargo test, and cargo bench are explicitly mentioned as validation steps.\n
- Context: While these are standard tools, cargo test and cargo bench execute code defined in the project, which could be malicious if the project source is untrusted.\n- [DYNAMIC_EXECUTION]: The skill utilizes tools that involve the compilation and execution of project-specific code at runtime.\n
- Evidence: Usage of cargo test and cargo bench.\n
- Context: These commands result in the execution of compiled test and benchmark binaries derived from the user's project.\n- [EXTERNAL_DOWNLOADS]: The skill refers to an external documentation site.\n
- Evidence: SKILL.md contains a link to https://jeffallan.github.io/claude-skills/skills/language/rust-engineer/.\n
- Note: This is a vendor-owned resource (GitHub Pages). Automated reputation scanners have flagged this URL, which should be considered when accessing the reference.
Recommendations
- CRITICAL: 1 file(s) identified as malware by FileRep - DO NOT USE
- Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata