akcli-deep-review
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill leverages the
akclicommand-line utility to perform hardware design analysis tasks, including schematic reading (akcli read), component inspection (akcli component), and the execution of a validation engine (akcli review validate). These operations are consistent with the skill's stated purpose of hardware design review. - [DATA_EXPOSURE]: The skill instructs the agent to process local engineering data, such as KiCad schematic files (
.kicad_sch) and datasheet facts. These files are accessed for the purpose of identifying design issues and are not transmitted to external or unauthorized domains. - [INDIRECT_PROMPT_INJECTION]: The skill handles data from external sources (schematic files and datasheets) which represent a potential indirect prompt injection surface. However, the skill provides significant mitigation by requiring all findings to be 'anchored' to existing design components and validated through a deterministic tool (
akcli review validate) before being accepted, preventing the agent from acting solely on potentially malicious instructions embedded in the design data.
Audit Metadata