review-llm-artifacts
Pass
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements legitimate functionality for a code review tool. It uses system commands (git, find, sed, jq) to identify files and manage analysis reports locally.
- [DATA_EXPOSURE_AND_EXFILTRATION]: No evidence of unauthorized data access or network transmission. The skill only identifies source code files for the purpose of analysis and writes findings to a local '.beagle' directory within the workspace.
- [COMMAND_EXECUTION]: Shell commands are used appropriately for file system navigation, git operations, and local JSON validation. Use of '$ARGUMENTS' follows standard pattern for agent skills to process user flags.
- [INDIRECT_PROMPT_INJECTION]: While the skill processes untrusted code files from the user's workspace, the capabilities triggered by this data are restricted to local analysis. The skill lacks dangerous outbound network capabilities or high-privilege execution paths that would make this surface critical.
Audit Metadata