keil
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
subprocess.runinscripts/keil_build.pyto invoke the Keil compiler (UV4.exe). This is the primary function of the skill and is implemented using list-based arguments, which prevents common shell injection vulnerabilities. The path to the executable and the projects are configurable via environment or project-level JSON files. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it reads and processes external data that could be influenced by an attacker.
- Ingestion points: The scripts
scripts/keil_project.pyandscripts/keil_build.pyread and parse Keil project files (.uvproj,.uvprojx) and build logs. - Boundary markers: None. The AI agent is instructed to review the content of these files for debugging MCU-specific issues like interrupt vectors or memory models.
- Capability inventory: The skill can execute shell commands via the Keil binary and perform file system writes for logs and configuration storage.
- Sanitization: The skill uses standard XML parsing and regex for log processing but does not include explicit instructions for the AI to disregard potentially malicious commands embedded within these files.
Audit Metadata