kmp-migration
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using Python to run local audit scripts. Evidence includes the execution of
audit_project.pyfrom.agents/skills/kmp-audit/and other local paths as documented in Phase 1 of SKILL.md. \n- [INDIRECT_PROMPT_INJECTION]: The skill defines a vulnerability surface by ingesting untrusted source code from an existing project to drive architectural changes. \n - Ingestion points: The audit scripts and checklists process all files in the project directory (
.), specifically reading ViewModels,settings.gradle.kts, and Gradle version catalogs. \n - Boundary markers: No explicit delimiters or instructions to ignore instructions within project data are present. \n
- Capability inventory: The skill facilitates shell execution (
python3), file system restructuring, and automated code rewriting (MVVM to MVI). \n - Sanitization: There are no mentioned mechanisms for sanitizing or validating the ingested project content before the agent acts upon it.
Audit Metadata