rust-syntax-unsafe
Warn
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: MEDIUMMETADATA_POISONINGINDIRECT_PROMPT_INJECTION
Full Analysis
- [METADATA_POISONING]: The metadata in the SKILL.md file lists 'OpenAEC-Foundation' as the author, which is inconsistent with the authoritative vendor identity 'impertio-studio'. This discrepancy is misleading regarding the skill's origin and maintenance.\n- [INDIRECT_PROMPT_INJECTION]: The skill's primary use case involves the agent reading and analyzing Rust source code provided by users, which serves as a potential ingestion point for malicious instructions.\n
- Ingestion points: Untrusted Rust source code snippets provided by the user for analysis or review in the agent's context.\n
- Boundary markers: While the skill strongly enforces the use of // SAFETY: comments for documenting invariants, these are intended for human review and do not function as programmatic security boundaries against prompt injection.\n
- Capability inventory: The skill instructions focus on code review logic and mention standard CLI tools like cargo and miri; no scripts for autonomous network exfiltration or unauthorized file system operations were identified.\n
- Sanitization: The skill does not include specific requirements for sanitizing or escaping the content of the user-provided code blocks before processing.
Audit Metadata