ai-native-assessment
Fail
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill's documentation includes a prerequisite step that fetches an installer for the Agentic Package Manager (APM) from a Microsoft-owned domain (
https://aka.ms/apm-unix). This is a reference to a well-known service and is documented neutrally. - [COMMAND_EXECUTION]: The workflow involves the execution of a bundled utility script,
scripts/detect.py, which performs a deterministic scan of the local repository to identify project architecture and configuration. This script operates locally and does not perform network activities. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. Since the assessment process requires the agent to read and evaluate the quality of source code, comments, and documentation within the target repository, malicious instructions could be embedded in those files to influence the agent's logic or the resulting maturity score.
- [SAFE]: The skill includes explicit security instructions that prohibit the inclusion of sensitive material (such as discovered hardcoded credentials or API keys) in the final scorecard. It requires the agent to use placeholders or redaction, demonstrating a security-by-design approach to reporting.
Recommendations
- HIGH: Downloads and executes remote code from: https://aka.ms/apm-unix - DO NOT USE without thorough review
Audit Metadata