dxe
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from target repository source files and runtime diagnostics to generate reports and trigger the loading of additional instruction sets.
- Ingestion points: Repository source code and platform-specific logs (e.g., browser console, Logcat) as defined in the Verification recipe.
- Boundary markers: No explicit delimiters are used to isolate untrusted data during processing or reporting.
- Capability inventory: File read/write access and the ability to execute repository-defined shell commands.
- Sanitization: No evidence of data sanitization or filtering of external source code or log content is provided.
- [DYNAMIC_EXECUTION]: The skill utilizes a 'Progressive sibling routing' system to load additional instruction sets from file paths computed at runtime (
../<skill-name>/SKILL.md) based on detected triggers. - [COMMAND_EXECUTION]: The 'Verification recipe' instructs the agent to execute shell commands from the target repository, such as existing typechecks, tests, and production builds, which results in the execution of local code.
Audit Metadata