mlir-development
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch a README file from the author's repository on GitHub (gmh5225/awesome-llvm-security) to obtain detailed resource links and tool lists.
- [INDIRECT_PROMPT_INJECTION]: The mechanism for fetching and processing external README content creates a vulnerability surface where remote data could influence agent behavior.
- Ingestion points:
SKILL.mdspecifies a remote URL (raw.githubusercontent.com) for data retrieval. - Boundary markers: The skill does not define specific delimiters or instructions to ignore potential commands within the fetched content.
- Capability inventory: The skill environment allows for execution of compiler tools like
mlir-optandmlir-translate. - Sanitization: There is no mention of sanitizing or validating the content retrieved from the remote URL.
Audit Metadata