asm-optimizer-artifacts
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists of instructional markdown and code examples for teaching assembly analysis. No malicious code, persistence mechanisms, or exfiltration patterns were detected.
- [COMMAND_EXECUTION]: Instructions provide standard GCC and diff commands for generating and comparing assembly files. These commands are limited to local file operations for analysis purposes.
- [INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface as it analyzes C source code. While no specific injection content is present, it is noted as a vulnerability surface.
- Ingestion points: C source code files processed by the agent in SKILL.md.
- Boundary markers: The skill instructs the agent to perform comparative analysis but lacks explicit markers to delimit untrusted source code.
- Capability inventory: The agent uses gcc, diff, and objdump (as described in SKILL.md and evals/README.md) to analyze files.
- Sanitization: The instructions do not provide specific sanitization or filtering for the source code being analyzed.
Audit Metadata