ets-runtime-dev
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill automates development tasks by executing shell commands, including git operations (branching, commits, diffs) and local build/test tools such as ark.py, ninja, es2abc, and ark_js_vm within the project environment.
- [DYNAMIC_EXECUTION]: The Planner agent dynamically constructs compilation and testing commands based on the current implementation plan and project state (e.g., detecting 'stub' files), using templates tailored for the ArkCompiler build system. This is a functional requirement for its build-validation workflow.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and analyze local source code and documentation, creating a surface for indirect prompt injection if those files contain malicious instructions.
- Ingestion points: Processes local source code (.cpp, .h, .ts), git diff outputs, and task tracking files (plan.md, tasks.md).
- Boundary markers: No specific delimiters or "ignore instructions" warnings are used when reading file contents for analysis or code review.
- Capability inventory: The skill can modify local files and execute shell commands (git, bash, and build scripts).
- Sanitization: Repository content is processed without explicit sanitization or filtering before being incorporated into the agent's context.
Audit Metadata